[
  {
    "id": "exo",
    "name": "Exo",
    "repo": "exo-explore/exo",
    "author": "exo-explore",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Decentralized Cluster Architects",
    "stars": 47769,
    "wGrowth": 158.2,
    "mGrowth": 125,
    "website": "https://github.com/exo-explore/exo",
    "traffic": "890k/mo",
    "searchVol": "160,000/mo",
    "latency": 24,
    "tokens": 91,
    "desc": "Run your own AI cluster at home with everyday devices (MacBooks, iPhones, iPads, PCs) connected in a p2p mesh.",
    "useCase": "Pool GPU memory from multiple personal MacBooks and iPhones to run massive 70B and 405B Llama models over local Wi-Fi.",
    "arch": "Python, Peer-to-peer ring topology, Dynamic tensor partitioning.",
    "tags": [
      "distributed-inference",
      "p2p-cluster",
      "apple-silicon-mesh",
      "exo"
    ],
    "starsDisplay": "47.8k",
    "weeklyStars": 34542,
    "weeklyGrowth": "+158.2%",
    "monthlyGrowth": "+125%",
    "wGrowthNum": 28.5,
    "mGrowthNum": 125,
    "trafficNum": 890000,
    "searchVolNum": 160000,
    "searchTrend": "+285%",
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "elo": 1421,
    "eloCi": "±18",
    "winRate": 95,
    "logoUrl": "https://github.com/exo-explore.png",
    "githubUrl": "https://github.com/exo-explore/exo",
    "rank": 1
  },
  {
    "id": "ktransformers",
    "name": "KTransformers",
    "repo": "kvcache-ai/ktransformers",
    "author": "kvcache-ai",
    "persona": "STEM & AI Researchers",
    "targetRole": "Kernel Heterogeneous Memory Architects",
    "stars": 19569,
    "wGrowth": 334.9,
    "mGrowth": 106,
    "website": "https://github.com/kvcache-ai/ktransformers",
    "traffic": "380k/mo",
    "searchVol": "66,000/mo",
    "latency": 15,
    "tokens": 96,
    "desc": "Flexible framework for ultra-low-cost, high-speed LLM inference offloading MoE weights to CPU RAM.",
    "useCase": "Run DeepSeek-V2/V3 236B MoE models on a single 24GB RTX 4090 GPU combined with standard DDR5 system RAM.",
    "arch": "C++, CUDA, CPU NUMA offloading, Custom AMX/AVX512 kernels.",
    "tags": [
      "moe-offloading",
      "deepseek-local",
      "hybrid-memory",
      "ktransformers"
    ],
    "starsDisplay": "19.6k",
    "weeklyStars": 16158,
    "weeklyGrowth": "+334.9%",
    "monthlyGrowth": "+106%",
    "wGrowthNum": 24.2,
    "mGrowthNum": 106,
    "trafficNum": 380000,
    "searchVolNum": 66000,
    "searchTrend": "+242%",
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1389,
    "eloCi": "±18",
    "winRate": 95,
    "logoUrl": "https://github.com/kvcache-ai.png",
    "githubUrl": "https://github.com/kvcache-ai/ktransformers",
    "rank": 2
  },
  {
    "id": "cogvideo",
    "name": "CogVideoX",
    "repo": "THUDM/CogVideo",
    "author": "THUDM",
    "persona": "STEM & AI Researchers",
    "targetRole": "Video Foundation Model Leads",
    "stars": 13063,
    "wGrowth": 76.5,
    "mGrowth": 96,
    "website": "https://github.com/THUDM/CogVideo",
    "traffic": "540k/mo",
    "searchVol": "95,000/mo",
    "latency": 32,
    "tokens": 86,
    "desc": "Open-source video generation foundation model series with 3D VAE and expert transformer architecture.",
    "useCase": "Generate cinematic 1080p high-frame-rate video sequences from text prompts and seed images locally.",
    "arch": "Python, PyTorch, 3D VAE, Diffusion Transformer (DiT).",
    "tags": [
      "video-foundation-model",
      "dit-video",
      "thudm",
      "generative-video"
    ],
    "starsDisplay": "13.1k",
    "weeklyStars": 7298,
    "weeklyGrowth": "+76.5%",
    "monthlyGrowth": "+96%",
    "wGrowthNum": 22.1,
    "mGrowthNum": 96,
    "trafficNum": 540000,
    "searchVolNum": 95000,
    "searchTrend": "+221%",
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "elo": 1375,
    "eloCi": "±18",
    "winRate": 95,
    "logoUrl": "https://github.com/THUDM.png",
    "githubUrl": "https://github.com/THUDM/CogVideo",
    "rank": 3
  },
  {
    "id": "docling",
    "name": "Docling",
    "repo": "DS4SD/docling",
    "author": "DS4SD",
    "persona": "STEM & AI Researchers",
    "targetRole": "Document Parsing Specialists",
    "stars": 68477,
    "wGrowth": 315,
    "mGrowth": 92,
    "website": "https://ds4sd.github.io/docling",
    "traffic": "680k/mo",
    "searchVol": "120,000/mo",
    "latency": 18,
    "tokens": 96,
    "desc": "Parse complex documents (PDF, DOCX, PPTX) into structured Markdown and JSON with table and formula recovery.",
    "useCase": "Extract high-density financial tables, charts, and mathematical equations from scientific whitepapers into LLM-clean formats.",
    "arch": "Python, DocLayNet models, TableFormer, OCR pipelines.",
    "tags": [
      "pdf-parser",
      "table-extraction",
      "document-ai",
      "docling"
    ],
    "starsDisplay": "68.5k",
    "weeklyStars": 55475,
    "weeklyGrowth": "+315%",
    "monthlyGrowth": "+92%",
    "wGrowthNum": 21.2,
    "mGrowthNum": 92,
    "trafficNum": 680000,
    "searchVolNum": 120000,
    "searchTrend": "+212%",
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1370,
    "eloCi": "±18",
    "winRate": 95,
    "logoUrl": "https://github.com/DS4SD.png",
    "githubUrl": "https://github.com/DS4SD/docling",
    "rank": 4
  },
  {
    "id": "mosec",
    "name": "Mosec",
    "repo": "mosecorg/mosec",
    "author": "mosecorg",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "High-Throughput Model Servers",
    "stars": 902,
    "starsDisplay": "902",
    "weeklyStars": 4215,
    "weeklyGrowth": "+21.2%",
    "monthlyGrowth": "+80%",
    "wGrowthNum": 21.2,
    "mGrowthNum": 80,
    "traffic": "459k/mo",
    "trafficNum": 459088,
    "searchVol": "115,000/mo",
    "searchVolNum": 114911,
    "searchTrend": "+212%",
    "latencyMs": "12ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 92,
    "elo": 1370,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/mosecorg/mosec",
    "githubUrl": "https://github.com/mosecorg/mosec",
    "logoUrl": "https://github.com/mosecorg.png",
    "desc": "High-impact open-source repository empowering high-throughput model servers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale rust-ipc infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "rust-ipc",
      "dynamic-batching",
      "pytorch-server",
      "mosec"
    ],
    "rank": 5
  },
  {
    "id": "triton-inference-server",
    "name": "Triton Inference Server",
    "repo": "triton-inference-server/server",
    "author": "triton-inference-server",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Production Inference Leads",
    "stars": 11048,
    "starsDisplay": "11.0k",
    "weeklyStars": 3714,
    "weeklyGrowth": "+20.7%",
    "monthlyGrowth": "+78%",
    "wGrowthNum": 20.7,
    "mGrowthNum": 78,
    "traffic": "777k/mo",
    "trafficNum": 777156,
    "searchVol": "93,000/mo",
    "searchVolNum": 92907,
    "searchTrend": "+207%",
    "latencyMs": "18ms",
    "latencyNum": 22,
    "tokensSaved": "-89%",
    "tokensNum": 95,
    "elo": 1366,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/triton-inference-server/server",
    "githubUrl": "https://github.com/triton-inference-server/server",
    "logoUrl": "https://github.com/triton-inference-server.png",
    "desc": "High-impact open-source repository empowering production inference leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale nvidia-triton infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "nvidia-triton",
      "model-serving",
      "multi-framework",
      "gpu-orchestration"
    ],
    "rank": 6
  },
  {
    "id": "fd",
    "name": "fd-find",
    "repo": "sharkdp/fd",
    "author": "sharkdp",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Filesystem Traversal Leads",
    "stars": 44652,
    "starsDisplay": "44.7k",
    "weeklyStars": 26970,
    "weeklyGrowth": "+100.8%",
    "monthlyGrowth": "+77%",
    "wGrowthNum": 20.5,
    "mGrowthNum": 77,
    "traffic": "374k/mo",
    "trafficNum": 373766,
    "searchVol": "140,000/mo",
    "searchVolNum": 139875,
    "searchTrend": "+205%",
    "latencyMs": "12ms",
    "latencyNum": 18,
    "tokensSaved": "-98%",
    "tokensNum": 93,
    "elo": 1365,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/sharkdp/fd",
    "githubUrl": "https://github.com/sharkdp/fd",
    "logoUrl": "https://github.com/sharkdp.png",
    "desc": "High-impact open-source tool empowering filesystem traversal leads with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical file-search infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "file-search",
      "rust-cli",
      "find-alternative",
      "filesystem"
    ],
    "rank": 7,
    "wGrowth": 100.8
  },
  {
    "id": "textgrad",
    "name": "TextGrad",
    "repo": "zou-group/textgrad",
    "author": "zou-group",
    "persona": "STEM & AI Researchers",
    "targetRole": "Text-Based Gradient Researchers",
    "stars": 3749,
    "starsDisplay": "3.7k",
    "weeklyStars": 7240,
    "weeklyGrowth": "+20.2%",
    "monthlyGrowth": "+76%",
    "wGrowthNum": 20.2,
    "mGrowthNum": 76,
    "traffic": "804k/mo",
    "trafficNum": 804473,
    "searchVol": "57,000/mo",
    "searchVolNum": 56879,
    "searchTrend": "+202%",
    "latencyMs": "26ms",
    "latencyNum": 17,
    "tokensSaved": "-98%",
    "tokensNum": 95,
    "elo": 1364,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/zou-group/textgrad",
    "githubUrl": "https://github.com/zou-group/textgrad",
    "logoUrl": "https://github.com/zou-group.png",
    "desc": "High-impact open-source repository empowering text-based gradient researchers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale text-gradients infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "text-gradients",
      "automatic-differentiation",
      "prompt-tuning",
      "stanford"
    ],
    "rank": 8
  },
  {
    "id": "leptos",
    "name": "Leptos",
    "repo": "leptos-rs/leptos",
    "author": "leptos-rs",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Full-Stack Rust Developers",
    "stars": 21369,
    "starsDisplay": "21.4k",
    "weeklyStars": 5274,
    "weeklyGrowth": "+20.1%",
    "monthlyGrowth": "+76%",
    "wGrowthNum": 20.1,
    "mGrowthNum": 76,
    "traffic": "699k/mo",
    "trafficNum": 699408,
    "searchVol": "97,000/mo",
    "searchVolNum": 96801,
    "searchTrend": "+201%",
    "latencyMs": "24ms",
    "latencyNum": 12,
    "tokensSaved": "-95%",
    "tokensNum": 90,
    "elo": 1363,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/leptos-rs/leptos",
    "githubUrl": "https://github.com/leptos-rs/leptos",
    "logoUrl": "https://github.com/leptos-rs.png",
    "desc": "High-impact open-source tool empowering full-stack rust developers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical rust-ui infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "rust-ui",
      "fine-grained-reactive",
      "ssr",
      "wasm"
    ],
    "rank": 9
  },
  {
    "id": "eza",
    "name": "eza",
    "repo": "eza-community/eza",
    "author": "eza-community",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Modern CLI Utility Specialists",
    "stars": 23488,
    "starsDisplay": "23.5k",
    "weeklyStars": 8989,
    "weeklyGrowth": "+29.6%",
    "monthlyGrowth": "+76%",
    "wGrowthNum": 20,
    "mGrowthNum": 76,
    "traffic": "964k/mo",
    "trafficNum": 964472,
    "searchVol": "67,000/mo",
    "searchVolNum": 67305,
    "searchTrend": "+200%",
    "latencyMs": "25ms",
    "latencyNum": 23,
    "tokensSaved": "-92%",
    "tokensNum": 90,
    "elo": 1361,
    "eloCi": "±18",
    "winRate": 95,
    "website": "https://github.com/eza-community/eza",
    "githubUrl": "https://github.com/eza-community/eza",
    "logoUrl": "https://github.com/eza-community.png",
    "desc": "High-impact open-source tool empowering modern cli utility specialists with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical ls-replacement infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "ls-replacement",
      "rust-cli",
      "git-status",
      "eza"
    ],
    "rank": 10,
    "wGrowth": 29.6
  },
  {
    "id": "bentoml",
    "name": "BentoML",
    "repo": "bentoml/BentoML",
    "author": "bentoml",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Model Serving Architects",
    "stars": 8882,
    "starsDisplay": "8.9k",
    "weeklyStars": 4023,
    "weeklyGrowth": "+19.9%",
    "monthlyGrowth": "+75%",
    "wGrowthNum": 19.9,
    "mGrowthNum": 75,
    "traffic": "409k/mo",
    "trafficNum": 408746,
    "searchVol": "54,000/mo",
    "searchVolNum": 54136,
    "searchTrend": "+199%",
    "latencyMs": "13ms",
    "latencyNum": 11,
    "tokensSaved": "-91%",
    "tokensNum": 98,
    "elo": 1361,
    "eloCi": "±18",
    "winRate": 94,
    "website": "https://github.com/bentoml/BentoML",
    "githubUrl": "https://github.com/bentoml/BentoML",
    "logoUrl": "https://github.com/bentoml.png",
    "desc": "High-impact open-source repository empowering model serving architects with production capabilities.",
    "useCase": "Used by modern agentic teams to scale model-packaging infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "model-packaging",
      "container-serving",
      "bentoml",
      "production-ai"
    ],
    "rank": 11
  },
  {
    "id": "mcp-proxy",
    "name": "MCP Proxy Gateway",
    "repo": "modelcontextprotocol/servers",
    "author": "modelcontextprotocol",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "MCP Gateway Engineers",
    "stars": 91056,
    "starsDisplay": "91.1k",
    "weeklyStars": 80137,
    "weeklyGrowth": "+569.6%",
    "monthlyGrowth": "+74%",
    "wGrowthNum": 19.7,
    "mGrowthNum": 74,
    "traffic": "435k/mo",
    "trafficNum": 434776,
    "searchVol": "103,000/mo",
    "searchVolNum": 103081,
    "searchTrend": "+197%",
    "latencyMs": "15ms",
    "latencyNum": 23,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1359,
    "eloCi": "±18",
    "winRate": 94,
    "website": "https://github.com/modelcontextprotocol/servers",
    "githubUrl": "https://github.com/modelcontextprotocol/servers",
    "logoUrl": "https://github.com/modelcontextprotocol.png",
    "desc": "High-impact open-source repository empowering mcp gateway engineers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale mcp-gateway infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "mcp-gateway",
      "protocol-proxy",
      "agent-mesh",
      "tool-server"
    ],
    "rank": 12,
    "wGrowth": 569.6
  },
  {
    "id": "open-notebooklm",
    "name": "Open NotebookLM",
    "repo": "gabrielchua/open-notebooklm",
    "author": "gabrielchua",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Synthesis Designers",
    "stars": 2598,
    "wGrowth": 19.5,
    "mGrowth": 95,
    "website": "https://github.com/gabrielchua/open-notebooklm",
    "traffic": "420k/mo",
    "searchVol": "75,000/mo",
    "latency": 45,
    "tokens": 88,
    "desc": "Open-source reproduction of Google NotebookLM podcast generation: convert documents into natural audio discussions.",
    "useCase": "Automatically transform research papers and PDFs into dynamic, two-speaker conversational podcast audio.",
    "arch": "Python, LangChain, Kokoro / Bark TTS, Streamlit.",
    "tags": [
      "podcast-generation",
      "notebooklm-clone",
      "audio-discussion",
      "tts"
    ],
    "starsDisplay": "2.6k",
    "weeklyStars": 1326,
    "weeklyGrowth": "+19.5%",
    "monthlyGrowth": "+95%",
    "wGrowthNum": 19.5,
    "mGrowthNum": 95,
    "trafficNum": 420000,
    "searchVolNum": 75000,
    "searchTrend": "+195%",
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "elo": 1357,
    "eloCi": "±18",
    "winRate": 94,
    "logoUrl": "https://github.com/gabrielchua.png",
    "githubUrl": "https://github.com/gabrielchua/open-notebooklm",
    "rank": 13
  },
  {
    "id": "litellm-proxy",
    "name": "LiteLLM Proxy",
    "repo": "BerriAI/litellm",
    "author": "BerriAI",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Universal API Gateway Specialists",
    "stars": 60262,
    "starsDisplay": "60.3k",
    "weeklyStars": 42017,
    "weeklyGrowth": "+166.2%",
    "monthlyGrowth": "+73%",
    "wGrowthNum": 19.4,
    "mGrowthNum": 73,
    "traffic": "1099k/mo",
    "trafficNum": 1099482,
    "searchVol": "136,000/mo",
    "searchVolNum": 136333,
    "searchTrend": "+194%",
    "latencyMs": "18ms",
    "latencyNum": 27,
    "tokensSaved": "-95%",
    "tokensNum": 88,
    "elo": 1358,
    "eloCi": "±18",
    "winRate": 94,
    "website": "https://github.com/BerriAI/litellm",
    "githubUrl": "https://github.com/BerriAI/litellm",
    "logoUrl": "https://github.com/BerriAI.png",
    "desc": "High-impact open-source repository empowering universal api gateway specialists with production capabilities.",
    "useCase": "Used by modern agentic teams to scale openai-proxy infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "openai-proxy",
      "rate-limiting",
      "load-balancer",
      "cost-tracking"
    ],
    "rank": 14,
    "wGrowth": 166.2
  },
  {
    "id": "deepeval",
    "name": "DeepEval",
    "repo": "confident-ai/deepeval",
    "author": "confident-ai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "LLM Unit Testing Specialists",
    "stars": 18671,
    "wGrowth": 201.1,
    "mGrowth": 85,
    "website": "https://confident-ai.com",
    "traffic": "440k/mo",
    "searchVol": "76,000/mo",
    "latency": 16,
    "tokens": 95,
    "desc": "The open-source LLM evaluation framework. Like Pytest, but for unit testing LLM applications.",
    "useCase": "Measure hallucination rate, answer relevancy, G-Eval score, and context recall with deterministic pass/fail thresholds in unit tests.",
    "arch": "Python, Pytest plugin, G-Eval methodology, Pydantic.",
    "tags": [
      "llm-evaluation",
      "unit-testing",
      "pytest-ai",
      "hallucination-metric"
    ],
    "starsDisplay": "18.7k",
    "weeklyStars": 13661,
    "weeklyGrowth": "+201.1%",
    "monthlyGrowth": "+85%",
    "wGrowthNum": 19.2,
    "mGrowthNum": 85,
    "trafficNum": 440000,
    "searchVolNum": 76000,
    "searchTrend": "+192%",
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "elo": 1355,
    "eloCi": "±18",
    "winRate": 94,
    "logoUrl": "https://github.com/confident-ai.png",
    "githubUrl": "https://github.com/confident-ai/deepeval",
    "rank": 15
  },
  {
    "id": "ripgrep",
    "name": "ripgrep",
    "repo": "BurntSushi/ripgrep",
    "author": "BurntSushi",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Regex Pattern Search Specialists",
    "stars": 68888,
    "starsDisplay": "68.9k",
    "weeklyStars": 43303,
    "weeklyGrowth": "+118.1%",
    "monthlyGrowth": "+72%",
    "wGrowthNum": 19,
    "mGrowthNum": 72,
    "traffic": "1051k/mo",
    "trafficNum": 1051455,
    "searchVol": "82,000/mo",
    "searchVolNum": 81554,
    "searchTrend": "+190%",
    "latencyMs": "18ms",
    "latencyNum": 9,
    "tokensSaved": "-98%",
    "tokensNum": 90,
    "elo": 1356,
    "eloCi": "±18",
    "winRate": 94,
    "website": "https://github.com/BurntSushi/ripgrep",
    "githubUrl": "https://github.com/BurntSushi/ripgrep",
    "logoUrl": "https://github.com/BurntSushi.png",
    "desc": "High-impact open-source tool empowering regex pattern search specialists with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical fast-regex infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "fast-regex",
      "rust-cli",
      "codebase-search",
      "ripgrep"
    ],
    "rank": 16,
    "wGrowth": 118.1
  },
  {
    "id": "cosyvoice",
    "name": "CosyVoice",
    "repo": "FunAudioLLM/CosyVoice",
    "author": "FunAudioLLM",
    "persona": "Music & Audio Creators",
    "targetRole": "Multilingual Speech Synthesis Leads",
    "stars": 23865,
    "wGrowth": 76.8,
    "mGrowth": 83,
    "website": "https://github.com/FunAudioLLM/CosyVoice",
    "traffic": "780k/mo",
    "searchVol": "135,000/mo",
    "latency": 24,
    "tokens": 93,
    "desc": "Multi-lingual large voice generation model by Alibaba FunAudioLLM for emotional voice control.",
    "useCase": "Control emotions, dialects, and cross-lingual voice cloning with fine-grained prosody guidance.",
    "arch": "Python, PyTorch, Flow-matching vocoder, Alibaba FunAudio.",
    "tags": [
      "emotional-tts",
      "cross-lingual",
      "alibaba-ai",
      "voice-synthesis"
    ],
    "starsDisplay": "23.9k",
    "weeklyStars": 12916,
    "weeklyGrowth": "+76.8%",
    "monthlyGrowth": "+83%",
    "wGrowthNum": 18.9,
    "mGrowthNum": 83,
    "trafficNum": 780000,
    "searchVolNum": 135000,
    "searchTrend": "+189%",
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1353,
    "eloCi": "±18",
    "winRate": 93,
    "logoUrl": "https://github.com/FunAudioLLM.png",
    "githubUrl": "https://github.com/FunAudioLLM/CosyVoice",
    "rank": 17
  },
  {
    "id": "embedchain",
    "name": "Embedchain",
    "repo": "mem0ai/embedchain",
    "author": "mem0ai",
    "persona": "Founders & Solo Creators",
    "targetRole": "Rapid RAG Builders",
    "stars": 14975,
    "starsDisplay": "15.0k",
    "weeklyStars": 2800,
    "weeklyGrowth": "+18.7%",
    "monthlyGrowth": "+71%",
    "wGrowthNum": 18.7,
    "mGrowthNum": 71,
    "traffic": "464k/mo",
    "trafficNum": 463628,
    "searchVol": "145,000/mo",
    "searchVolNum": 144829,
    "searchTrend": "+187%",
    "latencyMs": "25ms",
    "latencyNum": 29,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "elo": 1352,
    "eloCi": "±18",
    "winRate": 93,
    "website": "https://github.com/mem0ai/embedchain",
    "githubUrl": "https://github.com/mem0ai/embedchain",
    "logoUrl": "https://github.com/mem0ai.png",
    "desc": "High-impact open-source repository empowering rapid rag builders with production capabilities.",
    "useCase": "Used by modern agentic teams to scale data-to-bot infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "data-to-bot",
      "rapid-rag",
      "llm-wrapper",
      "mem0"
    ],
    "rank": 18
  },
  {
    "id": "tinygrad",
    "name": "tinygrad",
    "repo": "tinygrad/tinygrad",
    "author": "tinygrad",
    "persona": "STEM & AI Researchers",
    "targetRole": "Minimal Deep Learning Hackers",
    "stars": 33711,
    "starsDisplay": "33.7k",
    "weeklyStars": 23505,
    "weeklyGrowth": "+168.5%",
    "monthlyGrowth": "+71%",
    "wGrowthNum": 18.7,
    "mGrowthNum": 71,
    "traffic": "582k/mo",
    "trafficNum": 582444,
    "searchVol": "66,000/mo",
    "searchVolNum": 66225,
    "searchTrend": "+187%",
    "latencyMs": "13ms",
    "latencyNum": 14,
    "tokensSaved": "-90%",
    "tokensNum": 96,
    "elo": 1352,
    "eloCi": "±18",
    "winRate": 93,
    "website": "https://github.com/tinygrad/tinygrad",
    "githubUrl": "https://github.com/tinygrad/tinygrad",
    "logoUrl": "https://github.com/tinygrad.png",
    "desc": "High-impact open-source repository empowering minimal deep learning hackers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale minimal-deep-learning infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "minimal-deep-learning",
      "geohot",
      "compilers",
      "gpu-kernels"
    ],
    "rank": 19,
    "wGrowth": 168.5
  },
  {
    "id": "graphrag",
    "name": "GraphRAG",
    "repo": "microsoft/graphrag",
    "author": "microsoft",
    "persona": "STEM & AI Researchers",
    "targetRole": "Knowledge Graph Researchers",
    "stars": 36241,
    "wGrowth": 52.3,
    "mGrowth": 86,
    "website": "https://microsoft.github.io/graphrag",
    "traffic": "1.4M/mo",
    "searchVol": "250,000/mo",
    "latency": 35,
    "tokens": 82,
    "desc": "A modular graph-based Retrieval-Augmented Generation pipeline using knowledge graphs for complex reasoning.",
    "useCase": "Synthesize global answers across large document corpora where standard vector similarity search misses multi-hop relationships.",
    "arch": "Python, NetworkX, Leiden graph clustering, LLM extraction.",
    "tags": [
      "graph-rag",
      "knowledge-graph",
      "multi-hop-reasoning",
      "microsoft"
    ],
    "starsDisplay": "36.2k",
    "weeklyStars": 16844,
    "weeklyGrowth": "+52.3%",
    "monthlyGrowth": "+86%",
    "wGrowthNum": 18.5,
    "mGrowthNum": 86,
    "trafficNum": 14000000,
    "searchVolNum": 250000,
    "searchTrend": "+185%",
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-82%",
    "tokensNum": 82,
    "elo": 1351,
    "eloCi": "±18",
    "winRate": 93,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/graphrag",
    "rank": 20
  },
  {
    "id": "litdata",
    "name": "LitData",
    "repo": "Lightning-AI/litdata",
    "author": "Lightning-AI",
    "persona": "STEM & AI Researchers",
    "targetRole": "AI Data Streaming Leads",
    "stars": 615,
    "starsDisplay": "615",
    "weeklyStars": 3125,
    "weeklyGrowth": "+18.4%",
    "monthlyGrowth": "+69%",
    "wGrowthNum": 18.4,
    "mGrowthNum": 69,
    "traffic": "906k/mo",
    "trafficNum": 906340,
    "searchVol": "49,000/mo",
    "searchVolNum": 49410,
    "searchTrend": "+184%",
    "latencyMs": "15ms",
    "latencyNum": 26,
    "tokensSaved": "-89%",
    "tokensNum": 97,
    "elo": 1350,
    "eloCi": "±18",
    "winRate": 93,
    "website": "https://github.com/Lightning-AI/litdata",
    "githubUrl": "https://github.com/Lightning-AI/litdata",
    "logoUrl": "https://github.com/Lightning-AI.png",
    "desc": "High-impact open-source repository empowering ai data streaming leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale data-streaming infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "data-streaming",
      "pytorch-dataloader",
      "lightning-ai",
      "fast-io"
    ],
    "rank": 21
  },
  {
    "id": "llama-factory",
    "name": "LLaMA-Factory",
    "repo": "hiyouga/LLaMA-Factory",
    "author": "hiyouga",
    "persona": "STEM & AI Researchers",
    "targetRole": "Unified Fine-Tuning WebUI Specialists",
    "stars": 75341,
    "wGrowth": 90.7,
    "mGrowth": 81,
    "website": "https://github.com/hiyouga/LLaMA-Factory",
    "traffic": "1.8M/mo",
    "searchVol": "320,000/mo",
    "latency": 15,
    "tokens": 96,
    "desc": "Unified efficient fine-tuning of 100+ LLMs & VLMs with WebUI, supporting LoRA, QLoRA, and RLHF.",
    "useCase": "Supervise fine-tuning, DPO, and PPO through an interactive web dashboard with zero coding required.",
    "arch": "Python, Gradio WebUI, PEFT, DeepSpeed, vLLM export.",
    "tags": [
      "unified-finetuning",
      "webui-finetune",
      "rlhf",
      "llama-factory"
    ],
    "starsDisplay": "75.3k",
    "weeklyStars": 43030,
    "weeklyGrowth": "+90.7%",
    "monthlyGrowth": "+81%",
    "wGrowthNum": 18.2,
    "mGrowthNum": 81,
    "trafficNum": 18000000,
    "searchVolNum": 320000,
    "searchTrend": "+182%",
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1351,
    "eloCi": "±18",
    "winRate": 93,
    "logoUrl": "https://github.com/hiyouga.png",
    "githubUrl": "https://github.com/hiyouga/LLaMA-Factory",
    "rank": 22
  },
  {
    "id": "prefect",
    "name": "Prefect",
    "repo": "PrefectHQ/prefect",
    "author": "PrefectHQ",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Workflow Coordination Leads",
    "stars": 23982,
    "starsDisplay": "24.0k",
    "weeklyStars": 6612,
    "weeklyGrowth": "+18.1%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 18.1,
    "mGrowthNum": 68,
    "traffic": "1042k/mo",
    "trafficNum": 1042043,
    "searchVol": "123,000/mo",
    "searchVolNum": 123407,
    "searchTrend": "+181%",
    "latencyMs": "28ms",
    "latencyNum": 29,
    "tokensSaved": "-97%",
    "tokensNum": 89,
    "elo": 1350,
    "eloCi": "±18",
    "winRate": 93,
    "website": "https://github.com/PrefectHQ/prefect",
    "githubUrl": "https://github.com/PrefectHQ/prefect",
    "logoUrl": "https://github.com/PrefectHQ.png",
    "desc": "High-impact open-source repository empowering workflow coordination leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale workflow-orchestration infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "workflow-orchestration",
      "python-pipelines",
      "observability",
      "prefect"
    ],
    "rank": 23
  },
  {
    "id": "dsp",
    "name": "DSPy",
    "repo": "stanfordnlp/dspy",
    "author": "stanfordnlp",
    "persona": "STEM & AI Researchers",
    "targetRole": "Algorithmic Prompt Compilers",
    "stars": 38533,
    "starsDisplay": "38.5k",
    "weeklyStars": 26449,
    "weeklyGrowth": "+161.5%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 18,
    "mGrowthNum": 68,
    "traffic": "881k/mo",
    "trafficNum": 881107,
    "searchVol": "67,000/mo",
    "searchVolNum": 67072,
    "searchTrend": "+180%",
    "latencyMs": "22ms",
    "latencyNum": 21,
    "tokensSaved": "-94%",
    "tokensNum": 93,
    "elo": 1347,
    "eloCi": "±18",
    "winRate": 93,
    "website": "https://github.com/stanfordnlp/dspy",
    "githubUrl": "https://github.com/stanfordnlp/dspy",
    "logoUrl": "https://github.com/stanfordnlp.png",
    "desc": "High-impact open-source repository empowering algorithmic prompt compilers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale prompt-compiler infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "prompt-compiler",
      "teleprompter",
      "stanford-nlp",
      "dspy"
    ],
    "rank": 24,
    "wGrowth": 161.5
  },
  {
    "id": "zed",
    "name": "Zed Editor",
    "repo": "zed-industries/zed",
    "author": "zed-industries",
    "persona": "Autonomous Software Engineers",
    "targetRole": "High-Performance Editor Engineers",
    "stars": 91382,
    "starsDisplay": "91.4k",
    "weeklyStars": 67119,
    "weeklyGrowth": "+209.2%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 17.9,
    "mGrowthNum": 68,
    "traffic": "512k/mo",
    "trafficNum": 511941,
    "searchVol": "130,000/mo",
    "searchVolNum": 129693,
    "searchTrend": "+179%",
    "latencyMs": "15ms",
    "latencyNum": 24,
    "tokensSaved": "-91%",
    "tokensNum": 92,
    "elo": 1348,
    "eloCi": "±18",
    "winRate": 92,
    "website": "https://github.com/zed-industries/zed",
    "githubUrl": "https://github.com/zed-industries/zed",
    "logoUrl": "https://github.com/zed-industries.png",
    "desc": "High-impact open-source tool empowering high-performance editor engineers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical zed-editor infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "zed-editor",
      "gpui",
      "rust-ide",
      "collaborative-coding"
    ],
    "rank": 25,
    "wGrowth": 209.2
  },
  {
    "id": "open-hands",
    "name": "OpenHands",
    "repo": "All-Hands-AI/OpenHands",
    "author": "All-Hands-AI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous SWE & Systems Devs",
    "stars": 90143,
    "wGrowth": 93.9,
    "mGrowth": 79,
    "website": "https://all-hands.dev",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "190,000/mo",
    "searchVolNum": 190000,
    "searchTrend": "+260%",
    "latency": 30,
    "tokens": 87,
    "desc": "Autonomous software development agent platform (formerly OpenDevin) that writes code, fixes bugs, and ships PRs.",
    "useCase": "Run complete SWE-bench problem resolution workflows inside secure Docker sandboxes with event-driven agents.",
    "arch": "Python, Docker container sandbox, Event-driven architecture, React frontend.",
    "tags": [
      "autonomous-swe",
      "opendevin",
      "swe-bench",
      "docker-sandbox"
    ],
    "elo": 1345,
    "eloCi": "±16",
    "winRate": 90,
    "starsDisplay": "90.1k",
    "weeklyStars": 51641,
    "weeklyGrowth": "+93.9%",
    "monthlyGrowth": "+79%",
    "wGrowthNum": 17.2,
    "mGrowthNum": 79,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-87%",
    "tokensNum": 87,
    "logoUrl": "https://github.com/All-Hands-AI.png",
    "githubUrl": "https://github.com/All-Hands-AI/OpenHands",
    "rank": 26
  },
  {
    "id": "deno",
    "name": "Deno",
    "repo": "denoland/deno",
    "author": "denoland",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Secure TypeScript Runtime Architects",
    "stars": 108683,
    "starsDisplay": "108.7k",
    "weeklyStars": 92803,
    "weeklyGrowth": "+467.4%",
    "monthlyGrowth": "+64%",
    "wGrowthNum": 17.1,
    "mGrowthNum": 64,
    "traffic": "838k/mo",
    "trafficNum": 837743,
    "searchVol": "90,000/mo",
    "searchVolNum": 89762,
    "searchTrend": "+171%",
    "latencyMs": "17ms",
    "latencyNum": 24,
    "tokensSaved": "-92%",
    "tokensNum": 91,
    "elo": 1341,
    "eloCi": "±18",
    "winRate": 92,
    "website": "https://github.com/denoland/deno",
    "githubUrl": "https://github.com/denoland/deno",
    "logoUrl": "https://github.com/denoland.png",
    "desc": "High-impact open-source tool empowering secure typescript runtime architects with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical typescript-runtime infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "typescript-runtime",
      "v8",
      "web-standards",
      "security"
    ],
    "rank": 27,
    "wGrowth": 467.4
  },
  {
    "id": "agentops",
    "name": "AgentOps",
    "repo": "AgentOps-AI/agentops",
    "author": "AgentOps-AI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous SWE & Systems Devs",
    "stars": 5887,
    "wGrowth": 22.6,
    "mGrowth": 78,
    "website": "https://agentops.ai",
    "traffic": "190k/mo",
    "trafficNum": 190000,
    "searchVol": "42,000/mo",
    "searchVolNum": 42000,
    "searchTrend": "+240%",
    "latency": 18,
    "tokens": 94,
    "desc": "Python SDK for AI agent monitoring, observability, session replay, and deterministic regression testing.",
    "useCase": "Record step-by-step agent tool calls, API spend, and failure modes with visual time-travel debugging.",
    "arch": "Python, Session Tracing, REST telemetry, React dashboard.",
    "tags": [
      "agent-observability",
      "debugging",
      "telemetry",
      "ai-evaluation"
    ],
    "elo": 1275,
    "eloCi": "±20",
    "winRate": 83,
    "starsDisplay": "5.9k",
    "weeklyStars": 1893,
    "weeklyGrowth": "+22.6%",
    "monthlyGrowth": "+78%",
    "wGrowthNum": 16.8,
    "mGrowthNum": 78,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/AgentOps-AI.png",
    "githubUrl": "https://github.com/AgentOps-AI/agentops",
    "rank": 28
  },
  {
    "id": "lapce",
    "name": "Lapce",
    "repo": "lapce/lapce",
    "author": "lapce",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Native GUI Editor Developers",
    "stars": 38900,
    "starsDisplay": "38.9k",
    "weeklyStars": 16955,
    "weeklyGrowth": "+47.5%",
    "monthlyGrowth": "+63%",
    "wGrowthNum": 16.8,
    "mGrowthNum": 63,
    "traffic": "1063k/mo",
    "trafficNum": 1063366,
    "searchVol": "77,000/mo",
    "searchVolNum": 76717,
    "searchTrend": "+168%",
    "latencyMs": "8ms",
    "latencyNum": 21,
    "tokensSaved": "-95%",
    "tokensNum": 98,
    "elo": 1340,
    "eloCi": "±18",
    "winRate": 91,
    "website": "https://github.com/lapce/lapce",
    "githubUrl": "https://github.com/lapce/lapce",
    "logoUrl": "https://github.com/lapce.png",
    "desc": "High-impact open-source tool empowering native gui editor developers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical rust-ide infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "rust-ide",
      "druid-gui",
      "rope-science",
      "tree-sitter"
    ],
    "rank": 29,
    "wGrowth": 47.5
  },
  {
    "id": "copilotkit",
    "name": "CopilotKit",
    "repo": "CopilotKit/CopilotKit",
    "author": "CopilotKit",
    "persona": "Founders & Solo Creators",
    "targetRole": "Frontend AI Developers",
    "stars": 37792,
    "wGrowth": 117.2,
    "mGrowth": 74,
    "website": "https://copilotkit.ai",
    "traffic": "680k/mo",
    "trafficNum": 680000,
    "searchVol": "110,000/mo",
    "searchVolNum": 110000,
    "searchTrend": "+230%",
    "latency": 20,
    "tokens": 89,
    "desc": "Open-source React UI components and backend SDK for integrating in-app AI copilots, textareas, and generative UI.",
    "useCase": "Drop full AI copilot sidebars, custom canvas actions, and AI form autofill into React/Next.js web applications.",
    "arch": "TypeScript, React context hooks, LangGraph/LlamaIndex bridge, AG-UI.",
    "tags": [
      "in-app-copilot",
      "generative-ui",
      "react-ai",
      "nextjs-components"
    ],
    "elo": 1315,
    "eloCi": "±17",
    "winRate": 88,
    "starsDisplay": "37.8k",
    "weeklyStars": 23263,
    "weeklyGrowth": "+117.2%",
    "monthlyGrowth": "+74%",
    "wGrowthNum": 16.5,
    "mGrowthNum": 74,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/CopilotKit.png",
    "githubUrl": "https://github.com/CopilotKit/CopilotKit",
    "rank": 30
  },
  {
    "id": "parler-tts",
    "name": "Parler-TTS",
    "repo": "huggingface/parler-tts",
    "author": "huggingface",
    "persona": "Music & Audio Creators",
    "targetRole": "Controllable Audio Engineers",
    "stars": 5588,
    "wGrowth": 16.5,
    "mGrowth": 75,
    "website": "https://github.com/huggingface/parler-tts",
    "traffic": "380k/mo",
    "searchVol": "62,000/mo",
    "latency": 28,
    "tokens": 91,
    "desc": "Lightweight text-to-speech model that generates high-quality speech with complete natural language style control.",
    "useCase": "Control pitch, tone, pace, and background acoustics simply by writing a descriptive prompt.",
    "arch": "PyTorch, Transformers, DAC neural audio codec.",
    "tags": [
      "controllable-tts",
      "huggingface",
      "speech-synthesis",
      "prompt-driven"
    ],
    "starsDisplay": "5.6k",
    "weeklyStars": 1497,
    "weeklyGrowth": "+16.5%",
    "monthlyGrowth": "+75%",
    "wGrowthNum": 16.5,
    "mGrowthNum": 75,
    "trafficNum": 380000,
    "searchVolNum": 62000,
    "searchTrend": "+165%",
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 91,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/parler-tts",
    "rank": 31
  },
  {
    "id": "marker",
    "name": "Marker",
    "repo": "VikParuchuri/marker",
    "author": "VikParuchuri",
    "persona": "STEM & AI Researchers",
    "targetRole": "PDF OCR Engineers",
    "stars": 40231,
    "wGrowth": 133.9,
    "mGrowth": 72,
    "website": "https://github.com/VikParuchuri/marker",
    "traffic": "710k/mo",
    "searchVol": "130,000/mo",
    "latency": 19,
    "tokens": 95,
    "desc": "Convert PDF to markdown quickly with high accuracy, preserving formatting, math equations, and code blocks.",
    "useCase": "Batch-convert academic papers, technical documentation, and textbooks into clean Markdown for vector ingestion.",
    "arch": "Python, PyTorch, SurLayout, Nougat vision heuristics.",
    "tags": [
      "pdf-to-markdown",
      "math-extraction",
      "high-accuracy-ocr",
      "rag-prep"
    ],
    "starsDisplay": "40.2k",
    "weeklyStars": 25852,
    "weeklyGrowth": "+133.9%",
    "monthlyGrowth": "+72%",
    "wGrowthNum": 16.4,
    "mGrowthNum": 72,
    "trafficNum": 710000,
    "searchVolNum": 130000,
    "searchTrend": "+164%",
    "latencyMs": "19ms",
    "latencyNum": 19,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "elo": 1336,
    "eloCi": "±18",
    "winRate": 91,
    "logoUrl": "https://github.com/VikParuchuri.png",
    "githubUrl": "https://github.com/VikParuchuri/marker",
    "rank": 32
  },
  {
    "id": "sensevoice",
    "name": "SenseVoice",
    "repo": "FunAudioLLM/SenseVoice",
    "author": "FunAudioLLM",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Understanding & Emotion Leads",
    "stars": 9453,
    "wGrowth": 21.2,
    "mGrowth": 71,
    "website": "https://github.com/FunAudioLLM/SenseVoice",
    "traffic": "490k/mo",
    "searchVol": "84,000/mo",
    "latency": 12,
    "tokens": 98,
    "desc": "High-speed multi-lingual speech recognition and audio understanding with emotion detection.",
    "useCase": "Transcribe audio 15x faster than Whisper while detecting speaker sentiment, laughing, and background music.",
    "arch": "Python, Non-autoregressive Transformer, FunAudioLLM.",
    "tags": [
      "fast-transcription",
      "emotion-recognition",
      "sound-event-detection",
      "audio"
    ],
    "starsDisplay": "9.5k",
    "weeklyStars": 2932,
    "weeklyGrowth": "+21.2%",
    "monthlyGrowth": "+71%",
    "wGrowthNum": 16.4,
    "mGrowthNum": 71,
    "trafficNum": 490000,
    "searchVolNum": 84000,
    "searchTrend": "+164%",
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 91,
    "logoUrl": "https://github.com/FunAudioLLM.png",
    "githubUrl": "https://github.com/FunAudioLLM/SenseVoice",
    "rank": 33
  },
  {
    "id": "helix",
    "name": "Helix Editor",
    "repo": "helix-editor/helix",
    "author": "helix-editor",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Modal Text Editor Architects",
    "stars": 46491,
    "starsDisplay": "46.5k",
    "weeklyStars": 20296,
    "weeklyGrowth": "+48.4%",
    "monthlyGrowth": "+62%",
    "wGrowthNum": 16.4,
    "mGrowthNum": 62,
    "traffic": "1048k/mo",
    "trafficNum": 1047630,
    "searchVol": "69,000/mo",
    "searchVolNum": 69384,
    "searchTrend": "+164%",
    "latencyMs": "20ms",
    "latencyNum": 11,
    "tokensSaved": "-91%",
    "tokensNum": 97,
    "elo": 1337,
    "eloCi": "±18",
    "winRate": 91,
    "website": "https://github.com/helix-editor/helix",
    "githubUrl": "https://github.com/helix-editor/helix",
    "logoUrl": "https://github.com/helix-editor.png",
    "desc": "High-impact open-source tool empowering modal text editor architects with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical modal-editor infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "modal-editor",
      "rust-tui",
      "kakoune-grammar",
      "tree-sitter"
    ],
    "rank": 34,
    "wGrowth": 48.4
  },
  {
    "id": "swc",
    "name": "SWC",
    "repo": "swc-project/swc",
    "author": "swc-project",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Rust Compiler Specialists",
    "stars": 34211,
    "starsDisplay": "34.2k",
    "weeklyStars": 13273,
    "weeklyGrowth": "+36.8%",
    "monthlyGrowth": "+61%",
    "wGrowthNum": 16.3,
    "mGrowthNum": 61,
    "traffic": "1121k/mo",
    "trafficNum": 1120875,
    "searchVol": "101,000/mo",
    "searchVolNum": 100597,
    "searchTrend": "+163%",
    "latencyMs": "20ms",
    "latencyNum": 11,
    "tokensSaved": "-92%",
    "tokensNum": 95,
    "elo": 1336,
    "eloCi": "±18",
    "winRate": 91,
    "website": "https://github.com/swc-project/swc",
    "githubUrl": "https://github.com/swc-project/swc",
    "logoUrl": "https://github.com/swc-project.png",
    "desc": "High-impact open-source tool empowering rust compiler specialists with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical rust-compiler infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "rust-compiler",
      "babel-replacement",
      "fast-transpiler",
      "swc"
    ],
    "rank": 35,
    "wGrowth": 36.8
  },
  {
    "id": "sam-2",
    "name": "Segment Anything 2",
    "repo": "facebookresearch/segment-anything-2",
    "author": "facebookresearch",
    "persona": "STEM & AI Researchers",
    "targetRole": "Visual Foundation Model Devs",
    "stars": 19962,
    "wGrowth": 16.2,
    "mGrowth": 73,
    "website": "https://ai.meta.com/sam2",
    "traffic": "1.5M/mo",
    "searchVol": "260,000/mo",
    "latency": 22,
    "tokens": 93,
    "desc": "Meta's foundation model for promptable visual segmentation in images and real-time videos.",
    "useCase": "Track and segment arbitrary visual objects across video frames with zero-shot point or box prompts.",
    "arch": "PyTorch, Memory Attention, Streaming video segmentation.",
    "tags": [
      "video-segmentation",
      "meta-ai",
      "sam-2",
      "visual-ai"
    ],
    "starsDisplay": "20.0k",
    "weeklyStars": 4459,
    "weeklyGrowth": "+16.2%",
    "monthlyGrowth": "+73%",
    "wGrowthNum": 16.2,
    "mGrowthNum": 73,
    "trafficNum": 1000000,
    "searchVolNum": 260000,
    "searchTrend": "+162%",
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 91,
    "logoUrl": "https://github.com/facebookresearch.png",
    "githubUrl": "https://github.com/facebookresearch/segment-anything-2",
    "rank": 36
  },
  {
    "id": "llm-rs",
    "name": "llm-rs",
    "repo": "rustformers/llm",
    "author": "rustformers",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Rust Native LLM Engineers",
    "stars": 6154,
    "starsDisplay": "6.2k",
    "weeklyStars": 5390,
    "weeklyGrowth": "+16.2%",
    "monthlyGrowth": "+61%",
    "wGrowthNum": 16.2,
    "mGrowthNum": 61,
    "traffic": "1130k/mo",
    "trafficNum": 1130200,
    "searchVol": "120,000/mo",
    "searchVolNum": 120404,
    "searchTrend": "+162%",
    "latencyMs": "17ms",
    "latencyNum": 12,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "elo": 1336,
    "eloCi": "±18",
    "winRate": 91,
    "website": "https://github.com/rustformers/llm",
    "githubUrl": "https://github.com/rustformers/llm",
    "logoUrl": "https://github.com/rustformers.png",
    "desc": "High-impact open-source repository empowering rust native llm engineers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale rust-llm infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "rust-llm",
      "ggml-bindings",
      "zero-c-runtime",
      "cli"
    ],
    "rank": 37
  },
  {
    "id": "dioxus",
    "name": "Dioxus",
    "repo": "DioxusLabs/dioxus",
    "author": "DioxusLabs",
    "persona": "Founders & Solo Creators",
    "targetRole": "Multiplatform Rust UI Engineers",
    "stars": 39335,
    "starsDisplay": "39.3k",
    "weeklyStars": 13440,
    "weeklyGrowth": "+27.4%",
    "monthlyGrowth": "+61%",
    "wGrowthNum": 16.1,
    "mGrowthNum": 61,
    "traffic": "522k/mo",
    "trafficNum": 522066,
    "searchVol": "131,000/mo",
    "searchVolNum": 130882,
    "searchTrend": "+161%",
    "latencyMs": "22ms",
    "latencyNum": 8,
    "tokensSaved": "-95%",
    "tokensNum": 96,
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 91,
    "website": "https://github.com/DioxusLabs/dioxus",
    "githubUrl": "https://github.com/DioxusLabs/dioxus",
    "logoUrl": "https://github.com/DioxusLabs.png",
    "desc": "High-impact open-source tool empowering multiplatform rust ui engineers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical cross-platform infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "cross-platform",
      "rust-frontend",
      "mobile-desktop-web",
      "dioxus"
    ],
    "rank": 38,
    "wGrowth": 27.4
  },
  {
    "id": "agent-zero",
    "name": "Agent Zero",
    "repo": "frdel/agent-zero",
    "author": "frdel",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Full-Stack Software Engineers",
    "stars": 19384,
    "wGrowth": 97.8,
    "mGrowth": 71,
    "website": "https://github.com/frdel/agent-zero",
    "traffic": "310k/mo",
    "trafficNum": 310000,
    "searchVol": "55,000/mo",
    "searchVolNum": 55000,
    "searchTrend": "+215%",
    "latency": 26,
    "tokens": 92,
    "desc": "Dynamic, self-evolving autonomous AI framework that uses computer tools, terminal execution, and subagents organically.",
    "useCase": "Let an agent write its own custom tools on the fly and recruit subagents to solve complex multi-hour programming tasks.",
    "arch": "Python, Docker container sandbox, Dynamic tool synthesis, Terminal IPC.",
    "tags": [
      "self-evolving-agent",
      "subagent-swarm",
      "terminal-execution",
      "autonomous-coding"
    ],
    "elo": 1295,
    "eloCi": "±19",
    "winRate": 85,
    "starsDisplay": "19.4k",
    "weeklyStars": 11132,
    "weeklyGrowth": "+97.8%",
    "monthlyGrowth": "+71%",
    "wGrowthNum": 15.8,
    "mGrowthNum": 71,
    "latencyMs": "26ms",
    "latencyNum": 26,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/frdel.png",
    "githubUrl": "https://github.com/frdel/agent-zero",
    "rank": 39
  },
  {
    "id": "surya",
    "name": "Surya",
    "repo": "VikParuchuri/surya",
    "author": "VikParuchuri",
    "persona": "STEM & AI Researchers",
    "targetRole": "Multilingual OCR Specialists",
    "stars": 21469,
    "wGrowth": 73.1,
    "mGrowth": 69,
    "website": "https://github.com/VikParuchuri/surya",
    "traffic": "510k/mo",
    "searchVol": "92,000/mo",
    "latency": 16,
    "tokens": 96,
    "desc": "OCR, layout analysis, and reading order detection supporting 90+ languages with sub-second page latency.",
    "useCase": "Detect bounding boxes, columns, and reading sequence across complex non-Latin scripts and magazines.",
    "arch": "Python, PyTorch vision transformers, CTC decoding.",
    "tags": [
      "multilingual-ocr",
      "layout-analysis",
      "reading-order",
      "vision-ai"
    ],
    "starsDisplay": "21.5k",
    "weeklyStars": 11028,
    "weeklyGrowth": "+73.1%",
    "monthlyGrowth": "+69%",
    "wGrowthNum": 15.8,
    "mGrowthNum": 69,
    "trafficNum": 510000,
    "searchVolNum": 92000,
    "searchTrend": "+158%",
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1331,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/VikParuchuri.png",
    "githubUrl": "https://github.com/VikParuchuri/surya",
    "rank": 40
  },
  {
    "id": "kokoro-tts",
    "name": "Kokoro TTS",
    "repo": "hexgrad/kokoro",
    "author": "hexgrad",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Engineers & Podcasters",
    "stars": 9191,
    "wGrowth": 15.6,
    "mGrowth": 72,
    "website": "https://huggingface.co/hexgrad/Kokoro-82M",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "160,000/mo",
    "searchVolNum": 160000,
    "searchTrend": "+280%",
    "latency": 35,
    "tokens": 92,
    "desc": "82M parameter lightweight text-to-speech model delivering studio-quality natural voices with Apache-licensed weights.",
    "useCase": "Synthesize full podcast episodes and conversational audio tracks locally on CPU in real-time with zero cloud API cost.",
    "arch": "PyTorch, StyleTTS2 variant, Espeak-ng phonemizer.",
    "tags": [
      "text-to-speech",
      "lightweight-tts",
      "open-weights",
      "speech-synthesis"
    ],
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/hexgrad.png",
    "githubUrl": "https://github.com/hexgrad/kokoro",
    "starsDisplay": "9.2k",
    "weeklyStars": 2262,
    "weeklyGrowth": "+15.6%",
    "monthlyGrowth": "+72.0%",
    "wGrowthNum": 15.6,
    "mGrowthNum": 72,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 41
  },
  {
    "id": "vanna",
    "name": "Vanna",
    "repo": "vanna-ai/vanna",
    "author": "vanna-ai",
    "persona": "Founders & Solo Creators",
    "targetRole": "BI & Data Platform Engineers",
    "stars": 23818,
    "wGrowth": 86.1,
    "mGrowth": 66,
    "website": "https://vanna.ai",
    "traffic": "560k/mo",
    "trafficNum": 560000,
    "searchVol": "92,000/mo",
    "searchVolNum": 92000,
    "searchTrend": "+185%",
    "latency": 25,
    "tokens": 88,
    "desc": "Open-source Python RAG framework for SQL generation: chat with your database using natural language.",
    "useCase": "Enable business stakeholders to query Snowflake, BigQuery, and Postgres securely via natural language chat.",
    "arch": "Python, Vector store schema embeddings, SQL execution sandbox.",
    "tags": [
      "text-to-sql",
      "bi-copilot",
      "sql-generation",
      "database-agent"
    ],
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "starsDisplay": "23.8k",
    "weeklyStars": 13015,
    "weeklyGrowth": "+86.1%",
    "monthlyGrowth": "+66%",
    "wGrowthNum": 15.6,
    "mGrowthNum": 66,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/vanna-ai.png",
    "githubUrl": "https://github.com/vanna-ai/vanna",
    "rank": 42
  },
  {
    "id": "librechat",
    "name": "LibreChat",
    "repo": "danny-avila/LibreChat",
    "author": "danny-avila",
    "persona": "Founders & Solo Creators",
    "targetRole": "Omni-Model Chat Engineers",
    "stars": 45356,
    "wGrowth": 102.5,
    "mGrowth": 68,
    "website": "https://librechat.ai",
    "traffic": "1.4M/mo",
    "searchVol": "240,000/mo",
    "latency": 22,
    "tokens": 93,
    "desc": "Enhanced ChatGPT clone: support for all major LLM providers, multi-user accounts, file upload, and custom plugins.",
    "useCase": "Unify Anthropic, OpenAI, Mistral, and local models under a single enterprise interface with code interpreter and image gen.",
    "arch": "Node.js, Express, MongoDB, React, Tailwind CSS.",
    "tags": [
      "chatgpt-clone",
      "multi-provider",
      "code-interpreter",
      "librechat"
    ],
    "starsDisplay": "45.4k",
    "weeklyStars": 26450,
    "weeklyGrowth": "+102.5%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 15.6,
    "mGrowthNum": 68,
    "trafficNum": 14000000,
    "searchVolNum": 240000,
    "searchTrend": "+156%",
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1331,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/danny-avila.png",
    "githubUrl": "https://github.com/danny-avila/LibreChat",
    "rank": 43
  },
  {
    "id": "bat",
    "name": "bat",
    "repo": "sharkdp/bat",
    "author": "sharkdp",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Syntax Highlighting CLI Devs",
    "stars": 60692,
    "starsDisplay": "60.7k",
    "weeklyStars": 38539,
    "weeklyGrowth": "+131.2%",
    "monthlyGrowth": "+59%",
    "wGrowthNum": 15.6,
    "mGrowthNum": 59,
    "traffic": "586k/mo",
    "trafficNum": 585722,
    "searchVol": "140,000/mo",
    "searchVolNum": 139503,
    "searchTrend": "+156%",
    "latencyMs": "8ms",
    "latencyNum": 24,
    "tokensSaved": "-92%",
    "tokensNum": 94,
    "elo": 1331,
    "eloCi": "±18",
    "winRate": 90,
    "website": "https://github.com/sharkdp/bat",
    "githubUrl": "https://github.com/sharkdp/bat",
    "logoUrl": "https://github.com/sharkdp.png",
    "desc": "High-impact open-source tool empowering syntax highlighting cli devs with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical cat-clone infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "cat-clone",
      "syntax-highlighting",
      "git-integration",
      "rust"
    ],
    "rank": 44,
    "wGrowth": 131.2
  },
  {
    "id": "deepseek-r1",
    "name": "DeepSeek-R1",
    "repo": "deepseek-ai/DeepSeek-R1",
    "author": "deepseek-ai",
    "persona": "Researchers & STEM",
    "targetRole": "Mathematicians & STEM Researchers",
    "stars": 91923,
    "wGrowth": 15.5,
    "mGrowth": 68,
    "website": "https://deepseek.com",
    "traffic": "14.2M/mo",
    "trafficNum": 14200000,
    "searchVol": "620,000/mo",
    "searchVolNum": 620000,
    "searchTrend": "+195%",
    "latency": 85,
    "tokens": 92,
    "desc": "Frontier open-weights reasoning model trained with pure reinforcement learning (DeepSeek-R1-Zero). Excels at math, coding, and formal logic.",
    "useCase": "Solve complex Olympiad mathematics, verify topological proofs, and write multi-step Socratic explanations without hallucination.",
    "arch": "Multi-Head Latent Attention (MLA), DeepSeekMoE, DualPipe RL.",
    "tags": [
      "chain-of-thought",
      "reinforcement-learning",
      "math",
      "reasoning"
    ],
    "rank": 45,
    "elo": 1365,
    "eloCi": "±18",
    "winRate": 92,
    "starsDisplay": "91.9k",
    "weeklyStars": 22295,
    "weeklyGrowth": "+15.5%",
    "monthlyGrowth": "+68.0%",
    "wGrowthNum": 15.5,
    "mGrowthNum": 68,
    "latencyMs": "85ms",
    "latencyNum": 85,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/deepseek-ai.png",
    "githubUrl": "https://github.com/deepseek-ai/DeepSeek-R1"
  },
  {
    "id": "florence-2",
    "name": "Microsoft Florence-2",
    "repo": "microsoft/Florence-2",
    "author": "microsoft",
    "persona": "STEM & AI Researchers",
    "targetRole": "Vision-Language Foundation Devs",
    "stars": 6800,
    "wGrowth": 15.5,
    "mGrowth": 68,
    "website": "https://huggingface.co/microsoft/Florence-2-large",
    "traffic": "540k/mo",
    "searchVol": "92,000/mo",
    "latency": 18,
    "tokens": 95,
    "desc": "Advanced vision foundation model capable of handling captioning, detection, grounding, and OCR in one unified model.",
    "useCase": "Extract dense text, bounding boxes, and visual grounding coordinates from documents and UI screenshots.",
    "arch": "PyTorch, DaViT vision backbone, Transformer sequence-to-sequence.",
    "tags": [
      "vision-language",
      "ocr",
      "visual-grounding",
      "microsoft"
    ],
    "starsDisplay": "6.8k",
    "weeklyStars": 1054,
    "weeklyGrowth": "+15.5%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 15.5,
    "mGrowthNum": 68,
    "trafficNum": 540000,
    "searchVolNum": 92000,
    "searchTrend": "+155%",
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "elo": 1329,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/Florence-2",
    "rank": 46
  },
  {
    "id": "fastembed",
    "name": "FastEmbed",
    "repo": "qdrant/fastembed",
    "author": "qdrant",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Lightweight Embedding Devs",
    "stars": 3236,
    "wGrowth": 15.5,
    "mGrowth": 67,
    "website": "https://github.com/qdrant/fastembed",
    "traffic": "280k/mo",
    "searchVol": "48,000/mo",
    "latency": 8,
    "tokens": 99,
    "desc": "Fast, lightweight Python library for generating vector embeddings locally without PyTorch dependencies.",
    "useCase": "Generate sentence embeddings using ONNX Runtime directly inside edge microservices without installing 5GB PyTorch wheels.",
    "arch": "Python, ONNX Runtime, Hugging Face Tokenizers.",
    "tags": [
      "fast-embeddings",
      "onnx-inference",
      "zero-pytorch",
      "qdrant"
    ],
    "starsDisplay": "3.2k",
    "weeklyStars": 589,
    "weeklyGrowth": "+15.5%",
    "monthlyGrowth": "+67%",
    "wGrowthNum": 15.5,
    "mGrowthNum": 67,
    "trafficNum": 280000,
    "searchVolNum": 48000,
    "searchTrend": "+155%",
    "latencyMs": "8ms",
    "latencyNum": 8,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "elo": 1328,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/qdrant.png",
    "githubUrl": "https://github.com/qdrant/fastembed",
    "rank": 47
  },
  {
    "id": "ragflow",
    "name": "RAGFlow",
    "repo": "infiniflow/ragflow",
    "author": "infiniflow",
    "persona": "Founders & Solo Creators",
    "targetRole": "Full-Stack AI Engineers",
    "stars": 91758,
    "wGrowth": 182.3,
    "mGrowth": 64,
    "website": "https://ragflow.io",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "140,000/mo",
    "searchVolNum": 140000,
    "searchTrend": "+190%",
    "latency": 35,
    "tokens": 88,
    "desc": "Open-source RAG engine based on deep document understanding with fine-grained chunking templates.",
    "useCase": "Extract structured intelligence from complex PDFs, Excel sheets, and slides with zero hallucinations.",
    "arch": "Python, DeepDoc OCR, Elasticsearch, Infinity tensor engine.",
    "tags": [
      "rag",
      "deepdoc",
      "document-ai",
      "retrieval"
    ],
    "elo": 1315,
    "eloCi": "±16",
    "winRate": 88,
    "starsDisplay": "91.8k",
    "weeklyStars": 64263,
    "weeklyGrowth": "+182.3%",
    "monthlyGrowth": "+64%",
    "wGrowthNum": 15.4,
    "mGrowthNum": 64,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/infiniflow.png",
    "githubUrl": "https://github.com/infiniflow/ragflow",
    "rank": 48
  },
  {
    "id": "cursor-rules",
    "name": "Awesome Cursorrules",
    "repo": "PatrickJS/awesome-cursorrules",
    "author": "PatrickJS",
    "persona": "Autonomous Software Engineers",
    "targetRole": "AI-Augmented Devs",
    "stars": 40887,
    "wGrowth": 122.2,
    "mGrowth": 58.1,
    "website": "https://github.com/PatrickJS/awesome-cursorrules",
    "traffic": "920,000/mo",
    "trafficNum": 920000,
    "searchVol": "184,000/mo",
    "searchVolNum": 184000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Awesome Cursorrules powering production AI workloads.",
    "useCase": "Integrate Awesome Cursorrules into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1341,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "40.9k",
    "weeklyStars": 25302,
    "weeklyGrowth": "+122.2%",
    "monthlyGrowth": "+58.1%",
    "wGrowthNum": 15.3,
    "mGrowthNum": 58.1,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/PatrickJS.png",
    "githubUrl": "https://github.com/PatrickJS/awesome-cursorrules",
    "rank": 49
  },
  {
    "id": "mcp-servers",
    "name": "Anthropic MCP",
    "repo": "modelcontextprotocol/servers",
    "author": "modelcontextprotocol",
    "persona": "AI Ops & Tooling",
    "targetRole": "AI Architects & Ops",
    "stars": 91056,
    "wGrowth": 101.5,
    "mGrowth": 65,
    "website": "https://modelcontextprotocol.io",
    "traffic": "2.8M/mo",
    "trafficNum": 2800000,
    "searchVol": "210,000/mo",
    "searchVolNum": 210000,
    "searchTrend": "+180%",
    "latency": 35,
    "tokens": 97,
    "desc": "Reference implementations for Model Context Protocol (MCP) servers, establishing the open standard for connecting AI models to data sources and tools.",
    "useCase": "Plug local PostgreSQL databases, GitHub repositories, Slack workspaces, and filesystems into Claude and agentic CLI tools securely.",
    "arch": "JSON-RPC 2.0, Stdio & SSE transports, TypeScript, Python.",
    "tags": [
      "mcp",
      "anthropic",
      "tool-calling",
      "interoperability",
      "standard"
    ],
    "rank": 50,
    "elo": 1311,
    "eloCi": "±19",
    "winRate": 89,
    "starsDisplay": "91.1k",
    "weeklyStars": 52726,
    "weeklyGrowth": "+101.5%",
    "monthlyGrowth": "+65.0%",
    "wGrowthNum": 15.2,
    "mGrowthNum": 65,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/modelcontextprotocol.png",
    "githubUrl": "https://github.com/modelcontextprotocol/servers"
  },
  {
    "id": "qwen-2-5",
    "name": "Qwen 2.5",
    "repo": "QwenLM/Qwen2.5",
    "author": "QwenLM",
    "persona": "Researchers & STEM",
    "targetRole": "AI Researchers & Foundation Model Devs",
    "stars": 27660,
    "wGrowth": 15.2,
    "mGrowth": 66,
    "website": "https://qwenlm.github.io",
    "traffic": "3.6M/mo",
    "trafficNum": 3600000,
    "searchVol": "410,000/mo",
    "searchVolNum": 410000,
    "searchTrend": "+240%",
    "latency": 65,
    "tokens": 94,
    "desc": "Frontier open-weights foundational model series supporting 128k context and state-of-the-art coding and math benchmarks.",
    "useCase": "Self-host high-parameter multilingual reasoning and instruction following competing directly with proprietary models.",
    "arch": "Dense transformer, RoPE embeddings, GQA (Grouped Query Attention), SwiGLU.",
    "tags": [
      "frontier-weights",
      "multilingual",
      "coding-benchmark",
      "math"
    ],
    "elo": 1362,
    "eloCi": "±16",
    "winRate": 92,
    "logoUrl": "https://github.com/QwenLM.png",
    "githubUrl": "https://github.com/QwenLM/Qwen2.5",
    "starsDisplay": "27.7k",
    "weeklyStars": 4834,
    "weeklyGrowth": "+15.2%",
    "monthlyGrowth": "+66.0%",
    "wGrowthNum": 15.2,
    "mGrowthNum": 66,
    "latencyMs": "65ms",
    "latencyNum": 65,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "rank": 51
  },
  {
    "id": "jan",
    "name": "Jan",
    "repo": "janhq/jan",
    "author": "janhq",
    "persona": "Founders & Solo Creators",
    "targetRole": "Privacy & Edge AI Engineers",
    "stars": 44833,
    "wGrowth": 76.5,
    "mGrowth": 68,
    "website": "https://jan.ai",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "210,000/mo",
    "searchVolNum": 210000,
    "searchTrend": "+210%",
    "latency": 30,
    "tokens": 92,
    "desc": "Open-source, 100% offline alternative to ChatGPT that runs local models with a native desktop GUI and local API server.",
    "useCase": "Run DeepSeek, Llama 3, and Mistral models locally on MacBook or workstation with zero cloud telemetry.",
    "arch": "Tauri, TypeScript, Nitro C++ inference engine, local SQLite.",
    "tags": [
      "offline-llm",
      "local-ai",
      "desktop-gui",
      "privacy-first"
    ],
    "elo": 1310,
    "eloCi": "±17",
    "winRate": 87,
    "starsDisplay": "44.8k",
    "weeklyStars": 23294,
    "weeklyGrowth": "+76.5%",
    "monthlyGrowth": "+68%",
    "wGrowthNum": 15.2,
    "mGrowthNum": 68,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/janhq.png",
    "githubUrl": "https://github.com/janhq/jan",
    "rank": 52
  },
  {
    "id": "open-sora",
    "name": "Open-Sora",
    "repo": "hpcaitech/Open-Sora",
    "author": "hpcaitech",
    "persona": "STEM & AI Researchers",
    "targetRole": "Democratized Video Researchers",
    "stars": 29857,
    "wGrowth": 38.9,
    "mGrowth": 66,
    "website": "https://github.com/hpcaitech/Open-Sora",
    "traffic": "890k/mo",
    "searchVol": "150,000/mo",
    "latency": 30,
    "tokens": 87,
    "desc": "Democratizing efficient video production for all through open-source Sora reproduction and scaling laws.",
    "useCase": "Train and run large-scale video diffusion models with ColossalAI parallel infrastructure.",
    "arch": "Python, PyTorch, ColossalAI, STDiT architecture.",
    "tags": [
      "open-sora",
      "video-generation",
      "colossalai",
      "diffusion-transformer"
    ],
    "starsDisplay": "29.9k",
    "weeklyStars": 11625,
    "weeklyGrowth": "+38.9%",
    "monthlyGrowth": "+66%",
    "wGrowthNum": 15.2,
    "mGrowthNum": 66,
    "trafficNum": 890000,
    "searchVolNum": 150000,
    "searchTrend": "+152%",
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-87%",
    "tokensNum": 87,
    "elo": 1328,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/hpcaitech.png",
    "githubUrl": "https://github.com/hpcaitech/Open-Sora",
    "rank": 53
  },
  {
    "id": "repomix",
    "name": "Repomix",
    "repo": "yamadashy/repomix",
    "author": "yamadashy",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous SWE & Prompt Engineers",
    "stars": 28734,
    "wGrowth": 102.4,
    "mGrowth": 64,
    "website": "https://repomix.com",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "92,000/mo",
    "searchVolNum": 92000,
    "searchTrend": "+220%",
    "latency": 16,
    "tokens": 95,
    "desc": "Pack entire codebases into single AI-friendly Markdown files with tree hierarchies, token counters, and security filtering.",
    "useCase": "Feed entire complex GitHub repositories into Gemini 1.5 Pro or Claude 3.5 Sonnet context windows seamlessly.",
    "arch": "TypeScript, Tiktoken, Gitignore parsing, Secret scanner.",
    "tags": [
      "token-compression",
      "context-packing",
      "repo-digest",
      "prompt-engineering"
    ],
    "elo": 1330,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/yamadashy.png",
    "githubUrl": "https://github.com/yamadashy/repomix",
    "starsDisplay": "28.7k",
    "weeklyStars": 16678,
    "weeklyGrowth": "+102.4%",
    "monthlyGrowth": "+64.0%",
    "wGrowthNum": 15.1,
    "mGrowthNum": 64,
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "rank": 54
  },
  {
    "id": "claude-engineer",
    "name": "Claude Engineer",
    "repo": "Doriandarko/claude-engineer",
    "author": "Doriandarko",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Agentic SWE CLI Engineers",
    "stars": 11213,
    "wGrowth": 15.1,
    "mGrowth": 66,
    "website": "https://github.com/Doriandarko/claude-engineer",
    "traffic": "380k/mo",
    "searchVol": "65,000/mo",
    "latency": 20,
    "tokens": 93,
    "desc": "Interactive CLI assistant powered by Claude 3.5 Sonnet with direct filesystem read/write and command execution capabilities.",
    "useCase": "Perform rapid project refactorings, bash testing, and code generation with native Claude 3.5 Sonnet artifact tools.",
    "arch": "Python, Anthropic SDK, Rich CLI, Subprocess executor.",
    "tags": [
      "claude-cli",
      "terminal-agent",
      "sonnet-3-5",
      "swe-tools"
    ],
    "starsDisplay": "11.2k",
    "weeklyStars": 2383,
    "weeklyGrowth": "+15.1%",
    "monthlyGrowth": "+66%",
    "wGrowthNum": 15.1,
    "mGrowthNum": 66,
    "trafficNum": 380000,
    "searchVolNum": 65000,
    "searchTrend": "+151%",
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1326,
    "eloCi": "±18",
    "winRate": 90,
    "logoUrl": "https://github.com/Doriandarko.png",
    "githubUrl": "https://github.com/Doriandarko/claude-engineer",
    "rank": 55
  },
  {
    "id": "smollm-2",
    "name": "SmolLM2",
    "repo": "huggingface/smollm",
    "author": "huggingface",
    "persona": "Researchers & STEM",
    "targetRole": "On-Device AI Researchers",
    "stars": 3918,
    "wGrowth": 15,
    "mGrowth": 57,
    "website": "https://github.com/huggingface/smollm",
    "traffic": "575,000/mo",
    "trafficNum": 575000,
    "searchVol": "115,000/mo",
    "searchVolNum": 115000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool SmolLM2 powering production AI workloads.",
    "useCase": "Integrate SmolLM2 into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1340,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "3.9k",
    "weeklyStars": 1725,
    "weeklyGrowth": "+15.0%",
    "monthlyGrowth": "+57%",
    "wGrowthNum": 15,
    "mGrowthNum": 57,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/smollm",
    "rank": 56
  },
  {
    "id": "roo-cline",
    "name": "Roo Code",
    "repo": "RooVetGit/Roo-Cline",
    "author": "RooVetGit",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous SWE & Systems Devs",
    "stars": 24283,
    "wGrowth": 31.3,
    "mGrowth": 62,
    "website": "https://roocode.com",
    "traffic": "980k/mo",
    "trafficNum": 980000,
    "searchVol": "140,000/mo",
    "searchVolNum": 140000,
    "searchTrend": "+180%",
    "latency": 28,
    "tokens": 88,
    "desc": "Autonomous coding agent for VS Code with custom modes, deep architectural reasoning, and multi-model routing.",
    "useCase": "Orchestrate large-scale repository refactors and multi-file feature additions autonomously with deterministic terminal execution.",
    "arch": "TypeScript, VS Code Extension API, Tree-sitter AST, OpenRouter cascading.",
    "tags": [
      "autonomous-coding",
      "vscode",
      "agentic-ide",
      "multi-model"
    ],
    "elo": 1320,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/RooVetGit.png",
    "githubUrl": "https://github.com/RooVetGit/Roo-Cline",
    "starsDisplay": "24.3k",
    "weeklyStars": 8521,
    "weeklyGrowth": "+31.3%",
    "monthlyGrowth": "+62.0%",
    "wGrowthNum": 14.8,
    "mGrowthNum": 62,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 57
  },
  {
    "id": "f5-tts",
    "name": "F5-TTS",
    "repo": "SWivid/F5-TTS",
    "author": "SWivid",
    "persona": "Music & Audio Creators",
    "targetRole": "Voice Actors & Media Creators",
    "stars": 15347,
    "wGrowth": 19.9,
    "mGrowth": 65,
    "website": "https://github.com/SWivid/F5-TTS",
    "traffic": "520k/mo",
    "trafficNum": 520000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+230%",
    "latency": 42,
    "tokens": 86,
    "desc": "Fairytales non-autoregressive speech synthesis system with flow matching for zero-shot voice cloning.",
    "useCase": "Clone natural speech with only a 5-second reference audio snippet without duration predictors or pitch extraction.",
    "arch": "PyTorch, Flow Matching, DiT (Diffusion Transformer), Vocos vocoder.",
    "tags": [
      "voice-cloning",
      "flow-matching",
      "text-to-speech",
      "diffusion-audio"
    ],
    "elo": 1315,
    "eloCi": "±19",
    "winRate": 87,
    "logoUrl": "https://github.com/SWivid.png",
    "githubUrl": "https://github.com/SWivid/F5-TTS",
    "starsDisplay": "15.3k",
    "weeklyStars": 4441,
    "weeklyGrowth": "+19.9%",
    "monthlyGrowth": "+65.0%",
    "wGrowthNum": 14.8,
    "mGrowthNum": 65,
    "latencyMs": "42ms",
    "latencyNum": 42,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "rank": 58
  },
  {
    "id": "phoenix",
    "name": "Phoenix",
    "repo": "Arize-ai/phoenix",
    "author": "Arize-ai",
    "persona": "STEM & AI Researchers",
    "targetRole": "AI Evaluation Engineers",
    "stars": 11738,
    "wGrowth": 89.3,
    "mGrowth": 63,
    "website": "https://phoenix.arize.com",
    "traffic": "280k/mo",
    "trafficNum": 280000,
    "searchVol": "54,000/mo",
    "searchVolNum": 54000,
    "searchTrend": "+180%",
    "latency": 20,
    "tokens": 92,
    "desc": "AI observability platform providing evaluation, tracing, and dataset curation for LLM applications and agent chains.",
    "useCase": "Pinpoint RAG retrieval failures and hallucinated agent tool parameters with OpenTelemetry-standard traces.",
    "arch": "Python, OpenTelemetry spans, React dashboard, SQLite/PostgreSQL.",
    "tags": [
      "llm-observability",
      "rag-evaluation",
      "opentelemetry-tracing",
      "arize"
    ],
    "elo": 1285,
    "eloCi": "±19",
    "winRate": 84,
    "starsDisplay": "11.7k",
    "weeklyStars": 6456,
    "weeklyGrowth": "+89.3%",
    "monthlyGrowth": "+63%",
    "wGrowthNum": 14.8,
    "mGrowthNum": 63,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/Arize-ai.png",
    "githubUrl": "https://github.com/Arize-ai/phoenix",
    "rank": 59
  },
  {
    "id": "marqo",
    "name": "Marqo",
    "repo": "marqo-ai/marqo",
    "author": "marqo-ai",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Multimodal Vector Search Leads",
    "stars": 5031,
    "wGrowth": 14.8,
    "mGrowth": 64,
    "website": "https://marqo.ai",
    "traffic": "320k/mo",
    "searchVol": "55,000/mo",
    "latency": 18,
    "tokens": 94,
    "desc": "End-to-end multimodal vector search engine combining vector generation, storage, and retrieval.",
    "useCase": "Index images, text, and audio without pre-generating vectors; Marqo handles embedding and retrieval natively.",
    "arch": "Python, OpenSearch/Vespa backend, SBERT/CLIP models.",
    "tags": [
      "multimodal-search",
      "image-search",
      "vector-engine",
      "marqo"
    ],
    "starsDisplay": "5.0k",
    "weeklyStars": 1197,
    "weeklyGrowth": "+14.8%",
    "monthlyGrowth": "+64%",
    "wGrowthNum": 14.8,
    "mGrowthNum": 64,
    "trafficNum": 320000,
    "searchVolNum": 55000,
    "searchTrend": "+148%",
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1324,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/marqo-ai.png",
    "githubUrl": "https://github.com/marqo-ai/marqo",
    "rank": 60
  },
  {
    "id": "lobe-chat",
    "name": "Lobe Chat",
    "repo": "lobehub/lobe-chat",
    "author": "lobehub",
    "persona": "Founders & Solo Creators",
    "targetRole": "Modern UI/UX AI Designers",
    "stars": 83028,
    "wGrowth": 59.7,
    "mGrowth": 65,
    "website": "https://lobehub.com",
    "traffic": "2.6M/mo",
    "searchVol": "440,000/mo",
    "latency": 18,
    "tokens": 95,
    "desc": "Open-source, modern-design LLMs/multi-modal chat framework with plugin ecosystem and text-to-speech synthesis.",
    "useCase": "Deploy ultra-polished personal AI workspaces with custom agent market, visual artifacts, and mobile PWA support.",
    "arch": "Next.js, React, Zustand, Vercel AI SDK, Ant Design Mobile.",
    "tags": [
      "modern-ui",
      "pwa",
      "agent-market",
      "lobehub"
    ],
    "starsDisplay": "83.0k",
    "weeklyStars": 38724,
    "weeklyGrowth": "+59.7%",
    "monthlyGrowth": "+65%",
    "wGrowthNum": 14.8,
    "mGrowthNum": 65,
    "trafficNum": 26000000,
    "searchVolNum": 440000,
    "searchTrend": "+148%",
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "elo": 1328,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/lobehub.png",
    "githubUrl": "https://github.com/lobehub/lobe-chat",
    "rank": 61
  },
  {
    "id": "rspack",
    "name": "Rspack",
    "repo": "web-infra-dev/rspack",
    "author": "web-infra-dev",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Enterprise Rust Bundler Leads",
    "stars": 12935,
    "starsDisplay": "12.9k",
    "weeklyStars": 2315,
    "weeklyGrowth": "+14.7%",
    "monthlyGrowth": "+55%",
    "wGrowthNum": 14.7,
    "mGrowthNum": 55,
    "traffic": "873k/mo",
    "trafficNum": 872577,
    "searchVol": "143,000/mo",
    "searchVolNum": 143105,
    "searchTrend": "+147%",
    "latencyMs": "16ms",
    "latencyNum": 19,
    "tokensSaved": "-93%",
    "tokensNum": 91,
    "elo": 1324,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/web-infra-dev/rspack",
    "githubUrl": "https://github.com/web-infra-dev/rspack",
    "logoUrl": "https://github.com/web-infra-dev.png",
    "desc": "High-impact open-source tool empowering enterprise rust bundler leads with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical rspack infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "rspack",
      "webpack-compatible",
      "rust-bundler",
      "bytedance"
    ],
    "rank": 62
  },
  {
    "id": "unsloth",
    "name": "Unsloth",
    "repo": "unslothai/unsloth",
    "author": "unslothai",
    "persona": "Researchers & STEM",
    "targetRole": "Fine-Tuning & Model Engineers",
    "stars": 77287,
    "wGrowth": 219.4,
    "mGrowth": 61,
    "website": "https://unsloth.ai",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "210,000/mo",
    "searchVolNum": 210000,
    "searchTrend": "+195%",
    "latency": 25,
    "tokens": 92,
    "desc": "5x faster, 80% less memory fine-tuning for Llama 3, Mistral, Qwen, and DeepSeek. Custom Triton GPU kernels.",
    "useCase": "Fine-tune 70B parameter open models on single consumer GPUs without out-of-memory crashes or accuracy degradation.",
    "arch": "Python, Custom OpenAI Triton kernels, PyTorch, LoRA / QLoRA.",
    "tags": [
      "fine-tuning",
      "triton",
      "gpu-optimization",
      "lora"
    ],
    "elo": 1340,
    "eloCi": "±17",
    "winRate": 90,
    "logoUrl": "https://github.com/unslothai.png",
    "githubUrl": "https://github.com/unslothai/unsloth",
    "starsDisplay": "77.3k",
    "weeklyStars": 56596,
    "weeklyGrowth": "+219.4%",
    "monthlyGrowth": "+61.0%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 61,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 63
  },
  {
    "id": "helicone",
    "name": "Helicone",
    "repo": "Helicone/helicone",
    "author": "Helicone",
    "persona": "AI Ops & Tooling",
    "targetRole": "AI Ops & FinOps",
    "stars": 6203,
    "wGrowth": 26.6,
    "mGrowth": 55.1,
    "website": "https://github.com/Helicone/helicone",
    "traffic": "245,000/mo",
    "trafficNum": 245000,
    "searchVol": "49,000/mo",
    "searchVolNum": 49000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Helicone powering production AI workloads.",
    "useCase": "Integrate Helicone into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1338,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "6.2k",
    "weeklyStars": 2014,
    "weeklyGrowth": "+26.6%",
    "monthlyGrowth": "+55.1%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 55.1,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/Helicone.png",
    "githubUrl": "https://github.com/Helicone/helicone",
    "rank": 64
  },
  {
    "id": "shadcn-vue",
    "name": "shadcn-vue",
    "repo": "radix-vue/shadcn-vue",
    "author": "radix-vue",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Vue Engineers",
    "stars": 10669,
    "wGrowth": 44.2,
    "mGrowth": 55.1,
    "website": "https://github.com/radix-vue/shadcn-vue",
    "traffic": "370,000/mo",
    "trafficNum": 370000,
    "searchVol": "74,000/mo",
    "searchVolNum": 74000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool shadcn-vue powering production AI workloads.",
    "useCase": "Integrate shadcn-vue into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1338,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "10.7k",
    "weeklyStars": 4342,
    "weeklyGrowth": "+44.2%",
    "monthlyGrowth": "+55.1%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 55.1,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/radix-vue.png",
    "githubUrl": "https://github.com/radix-vue/shadcn-vue",
    "rank": 65
  },
  {
    "id": "letta",
    "name": "Letta",
    "repo": "letta-ai/letta",
    "author": "letta-ai",
    "persona": "STEM & AI Researchers",
    "targetRole": "Memory Systems Engineers",
    "stars": 25063,
    "wGrowth": 76.5,
    "mGrowth": 65,
    "website": "https://letta.com",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "78,000/mo",
    "searchVolNum": 78000,
    "searchTrend": "+175%",
    "latency": 32,
    "tokens": 89,
    "desc": "Memory-first agent platform (formerly MemGPT) that manages stateful memory hierarchies for long-running AI agents.",
    "useCase": "Deploy agents that remember user preferences and workspace facts across indefinite conversation sessions.",
    "arch": "Python, PostgreSQL, Vector recall, Tiered OS-style memory.",
    "tags": [
      "persistent-memory",
      "memgpt",
      "stateful-agents",
      "hierarchical-memory"
    ],
    "elo": 1305,
    "eloCi": "±18",
    "winRate": 86,
    "starsDisplay": "25.1k",
    "weeklyStars": 12922,
    "weeklyGrowth": "+76.5%",
    "monthlyGrowth": "+65%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 65,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/letta-ai.png",
    "githubUrl": "https://github.com/letta-ai/letta",
    "rank": 66
  },
  {
    "id": "animatediff",
    "name": "AnimateDiff",
    "repo": "guoyww/AnimateDiff",
    "author": "guoyww",
    "persona": "Music & Audio Creators",
    "targetRole": "Video Motion Specialists",
    "stars": 12272,
    "wGrowth": 14.5,
    "mGrowth": 63,
    "website": "https://github.com/guoyww/AnimateDiff",
    "traffic": "620k/mo",
    "searchVol": "110,000/mo",
    "latency": 28,
    "tokens": 89,
    "desc": "Animate your personalized text-to-image diffusion models with temporal motion modeling modules.",
    "useCase": "Transform personalized SD1.5 and SDXL checkpoints into smooth 60fps animated video clips without per-video training.",
    "arch": "Python, PyTorch, Motion Module attention layers.",
    "tags": [
      "video-generation",
      "motion-module",
      "animation",
      "diffusion"
    ],
    "starsDisplay": "12.3k",
    "weeklyStars": 3038,
    "weeklyGrowth": "+14.5%",
    "monthlyGrowth": "+63%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 63,
    "trafficNum": 620000,
    "searchVolNum": 110000,
    "searchTrend": "+145%",
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "elo": 1322,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/guoyww.png",
    "githubUrl": "https://github.com/guoyww/AnimateDiff",
    "rank": 67
  },
  {
    "id": "wasm-bindgen",
    "name": "wasm-bindgen",
    "repo": "rustwasm/wasm-bindgen",
    "author": "rustwasm",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "WASM Bridge Engineers",
    "stars": 9175,
    "starsDisplay": "9.2k",
    "weeklyStars": 3288,
    "weeklyGrowth": "+14.5%",
    "monthlyGrowth": "+55%",
    "wGrowthNum": 14.5,
    "mGrowthNum": 55,
    "traffic": "1107k/mo",
    "trafficNum": 1107307,
    "searchVol": "79,000/mo",
    "searchVolNum": 78997,
    "searchTrend": "+145%",
    "latencyMs": "18ms",
    "latencyNum": 25,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "elo": 1323,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/rustwasm/wasm-bindgen",
    "githubUrl": "https://github.com/rustwasm/wasm-bindgen",
    "logoUrl": "https://github.com/rustwasm.png",
    "desc": "High-impact open-source tool empowering wasm bridge engineers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical webassembly infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "webassembly",
      "rust",
      "wasm-bridge",
      "edge"
    ],
    "rank": 68
  },
  {
    "id": "surya-ocr",
    "name": "Surya Document Engine",
    "repo": "VikParuchuri/surya",
    "author": "VikParuchuri",
    "persona": "STEM & AI Researchers",
    "targetRole": "OCR Layout Analysis Leads",
    "stars": 21469,
    "starsDisplay": "21.5k",
    "weeklyStars": 6657,
    "weeklyGrowth": "+24.1%",
    "monthlyGrowth": "+54%",
    "wGrowthNum": 14.4,
    "mGrowthNum": 54,
    "traffic": "602k/mo",
    "trafficNum": 602092,
    "searchVol": "145,000/mo",
    "searchVolNum": 144578,
    "searchTrend": "+144%",
    "latencyMs": "21ms",
    "latencyNum": 10,
    "tokensSaved": "-94%",
    "tokensNum": 98,
    "elo": 1322,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/VikParuchuri/surya",
    "githubUrl": "https://github.com/VikParuchuri/surya",
    "logoUrl": "https://github.com/VikParuchuri.png",
    "desc": "High-impact open-source repository empowering ocr layout analysis leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale layout-parser infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "layout-parser",
      "ocr-detection",
      "reading-order",
      "multilingual"
    ],
    "rank": 69,
    "wGrowth": 24.1
  },
  {
    "id": "oxc",
    "name": "Oxc",
    "repo": "oxc-project/oxc",
    "author": "oxc-project",
    "persona": "Autonomous Software Engineers",
    "targetRole": "JavaScript Tooling Pioneers",
    "stars": 22960,
    "starsDisplay": "23.0k",
    "weeklyStars": 4208,
    "weeklyGrowth": "+14.3%",
    "monthlyGrowth": "+54%",
    "wGrowthNum": 14.3,
    "mGrowthNum": 54,
    "traffic": "901k/mo",
    "trafficNum": 901096,
    "searchVol": "127,000/mo",
    "searchVolNum": 126928,
    "searchTrend": "+143%",
    "latencyMs": "11ms",
    "latencyNum": 15,
    "tokensSaved": "-93%",
    "tokensNum": 91,
    "elo": 1323,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/oxc-project/oxc",
    "githubUrl": "https://github.com/oxc-project/oxc",
    "logoUrl": "https://github.com/oxc-project.png",
    "desc": "High-impact open-source tool empowering javascript tooling pioneers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical oxc-parser infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "oxc-parser",
      "rust-linter",
      "ast-minifier",
      "compiler"
    ],
    "rank": 70
  },
  {
    "id": "pydantic-ai",
    "name": "PydanticAI",
    "repo": "pydantic/pydantic-ai",
    "author": "pydantic",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Python Backend & AI Engineers",
    "stars": 20465,
    "wGrowth": 26.3,
    "mGrowth": 60,
    "website": "https://ai.pydantic.dev",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "160,000/mo",
    "searchVolNum": 160000,
    "searchTrend": "+210%",
    "latency": 22,
    "tokens": 90,
    "desc": "Python agent framework engineered for production by the makers of Pydantic. Built on pure types, validation, and structured outputs.",
    "useCase": "Build type-safe generative workflows where LLM responses validate directly against Pydantic models with automatic retries.",
    "arch": "Python 3.10+, Pydantic V2 core, AsyncIO, TypeGuard.",
    "tags": [
      "python",
      "type-safety",
      "agent-framework",
      "structured-output"
    ],
    "elo": 1340,
    "eloCi": "±16",
    "winRate": 90,
    "logoUrl": "https://github.com/pydantic.png",
    "githubUrl": "https://github.com/pydantic/pydantic-ai",
    "starsDisplay": "20.5k",
    "weeklyStars": 6565,
    "weeklyGrowth": "+26.3%",
    "monthlyGrowth": "+60.0%",
    "wGrowthNum": 14.2,
    "mGrowthNum": 60,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "rank": 71
  },
  {
    "id": "phi-4",
    "name": "Microsoft Phi-4",
    "repo": "microsoft/phi-4",
    "author": "microsoft",
    "persona": "Researchers & STEM",
    "targetRole": "STEM Researchers",
    "stars": 15200,
    "wGrowth": 14.2,
    "mGrowth": 54,
    "website": "https://github.com/microsoft/phi-4",
    "traffic": "760,000/mo",
    "trafficNum": 760000,
    "searchVol": "152,000/mo",
    "searchVolNum": 152000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Microsoft Phi-4 powering production AI workloads.",
    "useCase": "Integrate Microsoft Phi-4 into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1337,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "15.2k",
    "weeklyStars": 2158,
    "weeklyGrowth": "+14.2%",
    "monthlyGrowth": "+54%",
    "wGrowthNum": 14.2,
    "mGrowthNum": 54,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/phi-4",
    "rank": 72
  },
  {
    "id": "mentat",
    "name": "Mentat",
    "repo": "AbanteAI/mentat",
    "author": "AbanteAI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Terminal SWE Specialists",
    "stars": 6100,
    "wGrowth": 14.2,
    "mGrowth": 62,
    "website": "https://mentat.ai",
    "traffic": "210k/mo",
    "searchVol": "38,000/mo",
    "latency": 22,
    "tokens": 94,
    "desc": "The AI coding assistant that lives in your terminal and coordinates directly with your Git repo.",
    "useCase": "Query and modify code across multiple directories with conversational CLI commands and interactive diff confirmation.",
    "arch": "Python, GitPython, Tree-sitter AST, Streaming parser.",
    "tags": [
      "terminal-swe",
      "cli-agent",
      "git-integration",
      "python"
    ],
    "starsDisplay": "6.1k",
    "weeklyStars": 866,
    "weeklyGrowth": "+14.2%",
    "monthlyGrowth": "+62%",
    "wGrowthNum": 14.2,
    "mGrowthNum": 62,
    "trafficNum": 210000,
    "searchVolNum": 38000,
    "searchTrend": "+142%",
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1320,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/AbanteAI.png",
    "githubUrl": "https://github.com/AbanteAI/mentat",
    "rank": 73
  },
  {
    "id": "zenml",
    "name": "ZenML",
    "repo": "zenml-io/zenml",
    "author": "zenml-io",
    "persona": "STEM & AI Researchers",
    "targetRole": "MLOps Pipeline Engineers",
    "stars": 5607,
    "starsDisplay": "5.6k",
    "weeklyStars": 3311,
    "weeklyGrowth": "+14.2%",
    "monthlyGrowth": "+53%",
    "wGrowthNum": 14.2,
    "mGrowthNum": 53,
    "traffic": "1115k/mo",
    "trafficNum": 1114635,
    "searchVol": "145,000/mo",
    "searchVolNum": 144816,
    "searchTrend": "+142%",
    "latencyMs": "25ms",
    "latencyNum": 23,
    "tokensSaved": "-88%",
    "tokensNum": 89,
    "elo": 1321,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/zenml-io/zenml",
    "githubUrl": "https://github.com/zenml-io/zenml",
    "logoUrl": "https://github.com/zenml-io.png",
    "desc": "High-impact open-source repository empowering mlops pipeline engineers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale mlops infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "mlops",
      "production-pipelines",
      "cloud-agnostic",
      "zenml"
    ],
    "rank": 74
  },
  {
    "id": "ell",
    "name": "Ell",
    "repo": "MadcowD/ell",
    "author": "MadcowD",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Functional Prompt Engineers",
    "stars": 5849,
    "starsDisplay": "5.8k",
    "weeklyStars": 5205,
    "weeklyGrowth": "+14.2%",
    "monthlyGrowth": "+53%",
    "wGrowthNum": 14.2,
    "mGrowthNum": 53,
    "traffic": "378k/mo",
    "trafficNum": 377692,
    "searchVol": "76,000/mo",
    "searchVolNum": 76421,
    "searchTrend": "+142%",
    "latencyMs": "18ms",
    "latencyNum": 12,
    "tokensSaved": "-92%",
    "tokensNum": 97,
    "elo": 1323,
    "eloCi": "±18",
    "winRate": 89,
    "website": "https://github.com/MadcowD/ell",
    "githubUrl": "https://github.com/MadcowD/ell",
    "logoUrl": "https://github.com/MadcowD.png",
    "desc": "High-impact open-source repository empowering functional prompt engineers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale functional-prompting infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "functional-prompting",
      "prompt-versioning",
      "bytecode-inspector",
      "python"
    ],
    "rank": 75
  },
  {
    "id": "coolify",
    "name": "Coolify",
    "repo": "coollabsio/coolify",
    "author": "coollabsio",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Self-Hosting Indie Hackers",
    "stars": 62671,
    "wGrowth": 44.1,
    "mGrowth": 59,
    "website": "https://coolify.io",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "320,000/mo",
    "searchVolNum": 320000,
    "searchTrend": "+210%",
    "latency": 18,
    "tokens": 92,
    "desc": "Open-source, self-hostable Heroku / Netlify / Vercel alternative. Deploy apps, databases, and services to your own VPS.",
    "useCase": "Host 50+ micro-SaaS projects on single Hetzner/OVH VPS instances with automated SSL and zero platform fees.",
    "arch": "PHP / Laravel, Livewire, Docker Engine, Traefik proxy.",
    "tags": [
      "self-hosted",
      "paas",
      "vps-management",
      "docker"
    ],
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/coollabsio.png",
    "githubUrl": "https://github.com/coollabsio/coolify",
    "starsDisplay": "62.7k",
    "weeklyStars": 25304,
    "weeklyGrowth": "+44.1%",
    "monthlyGrowth": "+59.0%",
    "wGrowthNum": 14.1,
    "mGrowthNum": 59,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 76
  },
  {
    "id": "openpipe",
    "name": "OpenPipe",
    "repo": "openpipe/openpipe",
    "author": "openpipe",
    "persona": "AI Ops & Tooling",
    "targetRole": "Model Fine-Tuners",
    "stars": 2832,
    "wGrowth": 14.1,
    "mGrowth": 53.6,
    "website": "https://github.com/openpipe/openpipe",
    "traffic": "240,000/mo",
    "trafficNum": 240000,
    "searchVol": "48,000/mo",
    "searchVolNum": 48000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool OpenPipe powering production AI workloads.",
    "useCase": "Integrate OpenPipe into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1336,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "2.8k",
    "weeklyStars": 677,
    "weeklyGrowth": "+14.1%",
    "monthlyGrowth": "+53.6%",
    "wGrowthNum": 14.1,
    "mGrowthNum": 53.6,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/openpipe.png",
    "githubUrl": "https://github.com/openpipe/openpipe",
    "rank": 77
  },
  {
    "id": "torchtune",
    "name": "Torchtune",
    "repo": "pytorch/torchtune",
    "author": "pytorch",
    "persona": "STEM & AI Researchers",
    "targetRole": "PyTorch Fine-Tuning Engineers",
    "stars": 5811,
    "wGrowth": 14.1,
    "mGrowth": 60,
    "website": "https://pytorch.org/torchtune",
    "traffic": "340k/mo",
    "trafficNum": 340000,
    "searchVol": "58,000/mo",
    "searchVolNum": 58000,
    "searchTrend": "+170%",
    "latency": 24,
    "tokens": 91,
    "desc": "PyTorch-native library for easily authoring, fine-tuning and experimenting with LLMs using modular building blocks.",
    "useCase": "Fine-tune Llama 3 and Qwen models with LoRA and QLoRA recipes on single consumer GPUs or multi-node clusters.",
    "arch": "PyTorch 2.x, Fully Sharded Data Parallel (FSDP), TorchCompile.",
    "tags": [
      "pytorch-native",
      "llm-fine-tuning",
      "lora",
      "fsdp"
    ],
    "elo": 1290,
    "eloCi": "±19",
    "winRate": 85,
    "starsDisplay": "5.8k",
    "weeklyStars": 1001,
    "weeklyGrowth": "+14.1%",
    "monthlyGrowth": "+60%",
    "wGrowthNum": 14.1,
    "mGrowthNum": 60,
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/pytorch.png",
    "githubUrl": "https://github.com/pytorch/torchtune",
    "rank": 78
  },
  {
    "id": "kernel-memory",
    "name": "Kernel Memory",
    "repo": "microsoft/kernel-memory",
    "author": "microsoft",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Memory Architecture Engineers",
    "stars": 2233,
    "wGrowth": 14.1,
    "mGrowth": 61,
    "website": "https://microsoft.github.io/kernel-memory",
    "traffic": "290k/mo",
    "searchVol": "52,000/mo",
    "latency": 20,
    "tokens": 94,
    "desc": "Multi-modal RAG service that handles document extraction, chunking, embedding generation, and vector retrieval.",
    "useCase": "Ingest PDFs, Word docs, web pages, and audio into persistent memory pools with continuous asynchronous indexing.",
    "arch": "C#, Azure Cognitive Services, Vector DB abstraction.",
    "tags": [
      "rag-service",
      "document-pipeline",
      "enterprise-rag",
      "microsoft"
    ],
    "starsDisplay": "2.2k",
    "weeklyStars": 818,
    "weeklyGrowth": "+14.1%",
    "monthlyGrowth": "+61%",
    "wGrowthNum": 14.1,
    "mGrowthNum": 61,
    "trafficNum": 290000,
    "searchVolNum": 52000,
    "searchTrend": "+141%",
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1319,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/kernel-memory",
    "rank": 79
  },
  {
    "id": "autoawq",
    "name": "AutoAWQ",
    "repo": "casper-hansen/AutoAWQ",
    "author": "casper-hansen",
    "persona": "STEM & AI Researchers",
    "targetRole": "AWQ Quantization Specialists",
    "stars": 2346,
    "wGrowth": 14.1,
    "mGrowth": 61,
    "website": "https://github.com/casper-hansen/AutoAWQ",
    "traffic": "310k/mo",
    "searchVol": "54,000/mo",
    "latency": 10,
    "tokens": 98,
    "desc": "Easy-to-use package for 4-bit AWQ (Activation-aware Weight Quantization) with 3x speedup.",
    "useCase": "Quantize LLMs to 4-bit with zero perplexity degradation using activation-aware weight importance.",
    "arch": "Python, PyTorch, C++/CUDA GEMM kernels, Hugging Face.",
    "tags": [
      "4-bit-awq",
      "model-compression",
      "vllm-compatible",
      "quantization"
    ],
    "starsDisplay": "2.3k",
    "weeklyStars": 550,
    "weeklyGrowth": "+14.1%",
    "monthlyGrowth": "+61%",
    "wGrowthNum": 14.1,
    "mGrowthNum": 61,
    "trafficNum": 310000,
    "searchVolNum": 54000,
    "searchTrend": "+141%",
    "latencyMs": "10ms",
    "latencyNum": 10,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "elo": 1319,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/casper-hansen.png",
    "githubUrl": "https://github.com/casper-hansen/AutoAWQ",
    "rank": 80
  },
  {
    "id": "block-goose",
    "name": "Goose",
    "repo": "block/goose",
    "author": "block",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Full-Stack Software Engineers",
    "stars": 55024,
    "wGrowth": 343.7,
    "mGrowth": 58,
    "website": "https://block.github.io/goose",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "85,000/mo",
    "searchVolNum": 85000,
    "searchTrend": "+165%",
    "latency": 32,
    "tokens": 84,
    "desc": "Open-source extensible AI agent that automates complex engineering tasks and interacts with the host filesystem.",
    "useCase": "Run autonomous build, lint, and deploy workflows from the CLI with verified MCP server extensions.",
    "arch": "Rust, MCP protocol integration, Local IPC, Vector memory.",
    "tags": [
      "cli-agent",
      "rust",
      "mcp-client",
      "automation"
    ],
    "elo": 1295,
    "eloCi": "±20",
    "winRate": 85,
    "logoUrl": "https://github.com/block.png",
    "githubUrl": "https://github.com/block/goose",
    "starsDisplay": "55.0k",
    "weeklyStars": 44348,
    "weeklyGrowth": "+343.7%",
    "monthlyGrowth": "+58.0%",
    "wGrowthNum": 13.9,
    "mGrowthNum": 58,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-84%",
    "tokensNum": 84,
    "rank": 81
  },
  {
    "id": "agenta",
    "name": "Agenta",
    "repo": "Agenta-AI/agenta",
    "author": "Agenta-AI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "LLMOps & Prompt Engineers",
    "stars": 4814,
    "wGrowth": 50.4,
    "mGrowth": 59,
    "website": "https://agenta.ai",
    "traffic": "140k/mo",
    "trafficNum": 140000,
    "searchVol": "28,000/mo",
    "searchVolNum": 28000,
    "searchTrend": "+160%",
    "latency": 22,
    "tokens": 91,
    "desc": "Developer-first open-source LLMOps platform for prompt engineering, versioning, automated evaluation, and staging.",
    "useCase": "Collaboratively test prompt variations and custom LLM parameters against golden test suites in a unified web playground.",
    "arch": "FastAPI, Next.js, Docker runner, MongoDB.",
    "tags": [
      "llmops",
      "prompt-engineering",
      "evaluation-platform",
      "playground"
    ],
    "elo": 1260,
    "eloCi": "±22",
    "winRate": 81,
    "starsDisplay": "4.8k",
    "weeklyStars": 2059,
    "weeklyGrowth": "+50.4%",
    "monthlyGrowth": "+59%",
    "wGrowthNum": 13.9,
    "mGrowthNum": 59,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/Agenta-AI.png",
    "githubUrl": "https://github.com/Agenta-AI/agenta",
    "rank": 82
  },
  {
    "id": "depth-anything",
    "name": "Depth Anything",
    "repo": "LiheYoung/Depth-Anything",
    "author": "LiheYoung",
    "persona": "STEM & AI Researchers",
    "targetRole": "Spatial & 3D Vision Researchers",
    "stars": 8222,
    "wGrowth": 13.9,
    "mGrowth": 60,
    "website": "https://depth-anything.github.io",
    "traffic": "680k/mo",
    "searchVol": "110,000/mo",
    "latency": 20,
    "tokens": 94,
    "desc": "Highly robust monocular depth estimation foundation model for spatial AI and 3D visual perception.",
    "useCase": "Estimate precise depth maps from 2D photos for visual agents and 3D environment reconstruction.",
    "arch": "PyTorch, Vision Transformer (ViT), DPT head.",
    "tags": [
      "monocular-depth",
      "spatial-ai",
      "3d-vision",
      "vit"
    ],
    "starsDisplay": "8.2k",
    "weeklyStars": 1974,
    "weeklyGrowth": "+13.9%",
    "monthlyGrowth": "+60%",
    "wGrowthNum": 13.9,
    "mGrowthNum": 60,
    "trafficNum": 680000,
    "searchVolNum": 110000,
    "searchTrend": "+139%",
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1318,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/LiheYoung.png",
    "githubUrl": "https://github.com/LiheYoung/Depth-Anything",
    "rank": 83
  },
  {
    "id": "browser-use",
    "name": "Browser-Use",
    "repo": "browser-use/browser-use",
    "author": "browser-use",
    "persona": "AI Ops & Tooling",
    "targetRole": "AI Automation Engineers",
    "stars": 117326,
    "wGrowth": 297.7,
    "mGrowth": 60.5,
    "website": "https://browser-use.com",
    "traffic": "950k/mo",
    "trafficNum": 950000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+140%",
    "latency": 65,
    "tokens": 89,
    "desc": "Autonomous AI agent that interacts with web browsers using vision and DOM tree extraction.",
    "useCase": "Automate complex authenticated web forms, booking processes, and invoice downloads.",
    "arch": "Playwright, LangChain, Vision LLMs.",
    "tags": [
      "browser-agent",
      "vision",
      "automation"
    ],
    "rank": 84,
    "elo": 1269,
    "eloCi": "±20",
    "winRate": 83,
    "starsDisplay": "117.3k",
    "weeklyStars": 91897,
    "weeklyGrowth": "+297.7%",
    "monthlyGrowth": "+60.5%",
    "wGrowthNum": 13.8,
    "mGrowthNum": 60.5,
    "latencyMs": "65ms",
    "latencyNum": 65,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/browser-use.png",
    "githubUrl": "https://github.com/browser-use/browser-use"
  },
  {
    "id": "drizzle-orm",
    "name": "Drizzle ORM",
    "repo": "drizzle-team/drizzle-orm",
    "author": "drizzle-team",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Edge & Serverless Developers",
    "stars": 35970,
    "wGrowth": 13.8,
    "mGrowth": 56,
    "website": "https://orm.drizzle.team",
    "traffic": "2.1M/mo",
    "trafficNum": 2100000,
    "searchVol": "310,000/mo",
    "searchVolNum": 310000,
    "searchTrend": "+195%",
    "latency": 8,
    "tokens": 96,
    "desc": "Lightweight, ultra-fast headless TypeScript ORM. Zero dependencies, SQL-like syntax, and maximum edge efficiency.",
    "useCase": "Deploy database queries to Cloudflare Workers and V8 isolates with zero cold-start overhead.",
    "arch": "Pure TypeScript, Zero runtime dependencies, SQL-first compilation.",
    "tags": [
      "drizzle",
      "edge-orm",
      "typescript",
      "cloudflare-d1"
    ],
    "elo": 1345,
    "eloCi": "±17",
    "winRate": 91,
    "logoUrl": "https://github.com/drizzle-team.png",
    "githubUrl": "https://github.com/drizzle-team/drizzle-orm",
    "starsDisplay": "36.0k",
    "weeklyStars": 5368,
    "weeklyGrowth": "+13.8%",
    "monthlyGrowth": "+56.0%",
    "wGrowthNum": 13.8,
    "mGrowthNum": 56,
    "latencyMs": "8ms",
    "latencyNum": 8,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "rank": 85
  },
  {
    "id": "arize-phoenix",
    "name": "Arize Phoenix",
    "repo": "Arize-ai/phoenix",
    "author": "Arize-ai",
    "persona": "AI Ops & Tooling",
    "targetRole": "MLOps Engineers",
    "stars": 11738,
    "wGrowth": 80.6,
    "mGrowth": 52.4,
    "website": "https://github.com/Arize-ai/phoenix",
    "traffic": "325,000/mo",
    "trafficNum": 325000,
    "searchVol": "65,000/mo",
    "searchVolNum": 65000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Arize Phoenix powering production AI workloads.",
    "useCase": "Integrate Arize Phoenix into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1335,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "11.7k",
    "weeklyStars": 6135,
    "weeklyGrowth": "+80.6%",
    "monthlyGrowth": "+52.4%",
    "wGrowthNum": 13.8,
    "mGrowthNum": 52.4,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/Arize-ai.png",
    "githubUrl": "https://github.com/Arize-ai/phoenix",
    "rank": 86
  },
  {
    "id": "truelens",
    "name": "TruLens",
    "repo": "truera/truelens",
    "author": "truera",
    "persona": "Autonomous Software Engineers",
    "targetRole": "LLM Observability Engineers",
    "stars": 3400,
    "wGrowth": 13.8,
    "mGrowth": 59,
    "website": "https://truelens.org",
    "traffic": "290k/mo",
    "searchVol": "48,000/mo",
    "latency": 19,
    "tokens": 93,
    "desc": "Evaluation and tracking tool for LLM applications with feedback functions and visual dashboard.",
    "useCase": "Track feedback functions on input, context, and output to surface hallucinations across millions of user sessions.",
    "arch": "Python, Streamlit dashboard, SQLite/PostgreSQL storage.",
    "tags": [
      "llm-observability",
      "feedback-functions",
      "rag-tracking",
      "truera"
    ],
    "starsDisplay": "3.4k",
    "weeklyStars": 469,
    "weeklyGrowth": "+13.8%",
    "monthlyGrowth": "+59%",
    "wGrowthNum": 13.8,
    "mGrowthNum": 59,
    "trafficNum": 290000,
    "searchVolNum": 48000,
    "searchTrend": "+138%",
    "latencyMs": "19ms",
    "latencyNum": 19,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1316,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/truera.png",
    "githubUrl": "https://github.com/truera/truelens",
    "rank": 87
  },
  {
    "id": "gorules",
    "name": "GoRules Zen",
    "repo": "gorules/zen",
    "author": "gorules",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Business Rules Engine Architects",
    "stars": 2045,
    "starsDisplay": "2.0k",
    "weeklyStars": 4466,
    "weeklyGrowth": "+13.8%",
    "monthlyGrowth": "+52%",
    "wGrowthNum": 13.8,
    "mGrowthNum": 52,
    "traffic": "899k/mo",
    "trafficNum": 899202,
    "searchVol": "54,000/mo",
    "searchVolNum": 54452,
    "searchTrend": "+138%",
    "latencyMs": "24ms",
    "latencyNum": 10,
    "tokensSaved": "-92%",
    "tokensNum": 90,
    "elo": 1319,
    "eloCi": "±18",
    "winRate": 88,
    "website": "https://github.com/gorules/zen",
    "githubUrl": "https://github.com/gorules/zen",
    "logoUrl": "https://github.com/gorules.png",
    "desc": "High-impact open-source repository empowering business rules engine architects with production capabilities.",
    "useCase": "Used by modern agentic teams to scale decision-engine infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "decision-engine",
      "json-rules",
      "rust-evaluator",
      "business-logic"
    ],
    "rank": 88
  },
  {
    "id": "portkey",
    "name": "Portkey AI Gateway",
    "repo": "Portkey-AI/gateway",
    "author": "Portkey-AI",
    "persona": "AI Ops & Tooling",
    "targetRole": "Enterprise Gateway Ops",
    "stars": 13140,
    "wGrowth": 68.5,
    "mGrowth": 51.7,
    "website": "https://github.com/Portkey-AI/gateway",
    "traffic": "390,000/mo",
    "trafficNum": 390000,
    "searchVol": "78,000/mo",
    "searchVolNum": 78000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Portkey AI Gateway powering production AI workloads.",
    "useCase": "Integrate Portkey AI Gateway into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1334,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "13.1k",
    "weeklyStars": 6401,
    "weeklyGrowth": "+68.5%",
    "monthlyGrowth": "+51.7%",
    "wGrowthNum": 13.6,
    "mGrowthNum": 51.7,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/Portkey-AI.png",
    "githubUrl": "https://github.com/Portkey-AI/gateway",
    "rank": 89
  },
  {
    "id": "webarena",
    "name": "WebArena",
    "repo": "web-arena-x/webarena",
    "author": "web-arena-x",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous Web Benchmarkers",
    "stars": 1619,
    "starsDisplay": "1.6k",
    "weeklyStars": 4437,
    "weeklyGrowth": "+13.6%",
    "monthlyGrowth": "+51%",
    "wGrowthNum": 13.6,
    "mGrowthNum": 51,
    "traffic": "985k/mo",
    "trafficNum": 984808,
    "searchVol": "88,000/mo",
    "searchVolNum": 88255,
    "searchTrend": "+136%",
    "latencyMs": "14ms",
    "latencyNum": 26,
    "tokensSaved": "-93%",
    "tokensNum": 94,
    "elo": 1318,
    "eloCi": "±18",
    "winRate": 88,
    "website": "https://github.com/web-arena-x/webarena",
    "githubUrl": "https://github.com/web-arena-x/webarena",
    "logoUrl": "https://github.com/web-arena-x.png",
    "desc": "High-impact open-source repository empowering autonomous web benchmarkers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale webarena infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "webarena",
      "end-to-end-benchmarks",
      "carnegie-mellon",
      "agent-eval"
    ],
    "rank": 90
  },
  {
    "id": "karpathy-autoresearch",
    "name": "AutoResearch",
    "repo": "karpathy/autoresearch",
    "author": "karpathy",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Staff AI Engineers",
    "stars": 97427,
    "wGrowth": 155,
    "mGrowth": 58,
    "website": "https://github.com/karpathy/autoresearch",
    "traffic": "1.9M/mo",
    "trafficNum": 1900000,
    "searchVol": "72,000/mo",
    "searchVolNum": 72000,
    "searchTrend": "+145%",
    "latency": 48,
    "tokens": 95,
    "desc": "Andrej Karpathy's autonomous agent loop that proposes, tests, verifies, and iterates on research hypotheses and codebases.",
    "useCase": "Automate overnight hyperparameter sweeps, test-driven refactorings, and benchmark verification loops without human intervention.",
    "arch": "Python, Minimalist LLM API harness, Git diff parser.",
    "tags": [
      "karpathy",
      "autonomous-coding",
      "research-loop",
      "minimalist"
    ],
    "rank": 91,
    "elo": 1284,
    "eloCi": "±21",
    "winRate": 85,
    "starsDisplay": "97.4k",
    "weeklyStars": 64384,
    "weeklyGrowth": "+155%",
    "monthlyGrowth": "+58.0%",
    "wGrowthNum": 13.5,
    "mGrowthNum": 58,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/karpathy.png",
    "githubUrl": "https://github.com/karpathy/autoresearch"
  },
  {
    "id": "fasthtml",
    "name": "FastHTML",
    "repo": "AnswerDotAI/fasthtml",
    "author": "AnswerDotAI",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Python Full-Stack Founders",
    "stars": 7052,
    "wGrowth": 13.5,
    "mGrowth": 54,
    "website": "https://fastht.ml",
    "traffic": "620k/mo",
    "trafficNum": 620000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+190%",
    "latency": 18,
    "tokens": 92,
    "desc": "Fast, modern web applications in pure Python without JavaScript build steps. First-class HTMX integration.",
    "useCase": "Ship hyper-fast agentic dashboards, admin interfaces, and real-time LLM streaming interfaces in 100% Python.",
    "arch": "Python, HTMX, Starlette, Uvicorn, ASGI.",
    "tags": [
      "python-web",
      "htmx",
      "full-stack",
      "fast-iteration"
    ],
    "elo": 1280,
    "eloCi": "±22",
    "winRate": 83,
    "logoUrl": "https://github.com/AnswerDotAI.png",
    "githubUrl": "https://github.com/AnswerDotAI/fasthtml",
    "starsDisplay": "7.1k",
    "weeklyStars": 1593,
    "weeklyGrowth": "+13.5%",
    "monthlyGrowth": "+54.0%",
    "wGrowthNum": 13.5,
    "mGrowthNum": 54,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 92
  },
  {
    "id": "mesop",
    "name": "Mesop",
    "repo": "google/mesop",
    "author": "google",
    "persona": "Founders & Solo Creators",
    "targetRole": "Rapid Prototype Engineers",
    "stars": 6586,
    "wGrowth": 13.5,
    "mGrowth": 59,
    "website": "https://google.github.io/mesop",
    "traffic": "320k/mo",
    "trafficNum": 320000,
    "searchVol": "62,000/mo",
    "searchVolNum": 62000,
    "searchTrend": "+160%",
    "latency": 24,
    "tokens": 88,
    "desc": "Google's Python-based UI framework that enables developers to rapidly build web UIs like demos and internal tools.",
    "useCase": "Create composable AI workflows and internal operations panels with Angular-grade components in Python.",
    "arch": "Python, Angular web components, State diffing, Hot reload.",
    "tags": [
      "python-web",
      "internal-tools",
      "google-open-source",
      "ui-framework"
    ],
    "elo": 1270,
    "eloCi": "±20",
    "winRate": 82,
    "starsDisplay": "6.6k",
    "weeklyStars": 1026,
    "weeklyGrowth": "+13.5%",
    "monthlyGrowth": "+59%",
    "wGrowthNum": 13.5,
    "mGrowthNum": 59,
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/google.png",
    "githubUrl": "https://github.com/google/mesop",
    "rank": 93
  },
  {
    "id": "magentic",
    "name": "Magentic",
    "repo": "jackmpcollins/magentic",
    "author": "jackmpcollins",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Typed Prompt Engineers",
    "stars": 2426,
    "wGrowth": 13.5,
    "mGrowth": 58,
    "website": "https://github.com/jackmpcollins/magentic",
    "traffic": "140k/mo",
    "searchVol": "28,000/mo",
    "latency": 15,
    "tokens": 96,
    "desc": "Seamlessly integrate LLMs as Python functions with Pydantic type signatures and structured validation.",
    "useCase": "Define Python functions decorated with `@prompt` that return type-safe Pydantic objects from LLM outputs.",
    "arch": "Python, Pydantic v2, Instructor primitives, Typing.",
    "tags": [
      "typed-prompts",
      "pydantic-llm",
      "structured-output",
      "pythonic"
    ],
    "starsDisplay": "2.4k",
    "weeklyStars": 527,
    "weeklyGrowth": "+13.5%",
    "monthlyGrowth": "+58%",
    "wGrowthNum": 13.5,
    "mGrowthNum": 58,
    "trafficNum": 140000,
    "searchVolNum": 28000,
    "searchTrend": "+135%",
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1314,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/jackmpcollins.png",
    "githubUrl": "https://github.com/jackmpcollins/magentic",
    "rank": 94
  },
  {
    "id": "moondream",
    "name": "Moondream2",
    "repo": "vikhyat/moondream",
    "author": "vikhyat",
    "persona": "Researchers & STEM",
    "targetRole": "Vision Researchers",
    "stars": 10095,
    "wGrowth": 13.4,
    "mGrowth": 50.9,
    "website": "https://github.com/vikhyat/moondream",
    "traffic": "520,000/mo",
    "trafficNum": 520000,
    "searchVol": "104,000/mo",
    "searchVolNum": 104000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Moondream2 powering production AI workloads.",
    "useCase": "Integrate Moondream2 into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1334,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "10.1k",
    "weeklyStars": 1394,
    "weeklyGrowth": "+13.4%",
    "monthlyGrowth": "+50.9%",
    "wGrowthNum": 13.4,
    "mGrowthNum": 50.9,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/vikhyat.png",
    "githubUrl": "https://github.com/vikhyat/moondream",
    "rank": 95
  },
  {
    "id": "lancedb",
    "name": "LanceDB",
    "repo": "lancedb/lancedb",
    "author": "lancedb",
    "persona": "AI Ops & Tooling",
    "targetRole": "Embedded & Edge AI Engineers",
    "stars": 11612,
    "wGrowth": 81.4,
    "mGrowth": 52,
    "website": "https://lancedb.com",
    "traffic": "420k/mo",
    "trafficNum": 420000,
    "searchVol": "75,000/mo",
    "searchVolNum": 75000,
    "searchTrend": "+180%",
    "latency": 12,
    "tokens": 95,
    "desc": "Serverless vector database for AI apps. Disk-backed columnar format designed for fast vector search on NVMe drives.",
    "useCase": "Embed billion-scale vector indexes locally without paying recurring cloud RAM instance costs.",
    "arch": "Rust, Lance columnar format, Apache Arrow, SIMD indexing.",
    "tags": [
      "serverless-vector",
      "rust",
      "columnar",
      "low-cost-search"
    ],
    "elo": 1290,
    "eloCi": "±21",
    "winRate": 84,
    "logoUrl": "https://github.com/lancedb.png",
    "githubUrl": "https://github.com/lancedb/lancedb",
    "starsDisplay": "11.6k",
    "weeklyStars": 6057,
    "weeklyGrowth": "+81.4%",
    "monthlyGrowth": "+52.0%",
    "wGrowthNum": 13.2,
    "mGrowthNum": 52,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "rank": 96
  },
  {
    "id": "mlx",
    "name": "Apple MLX",
    "repo": "ml-explore/mlx",
    "author": "ml-explore",
    "persona": "STEM & AI Researchers",
    "targetRole": "Apple Silicon AI Researchers",
    "stars": 28674,
    "wGrowth": 39.9,
    "mGrowth": 57,
    "website": "https://ml-explore.github.io/mlx",
    "traffic": "720k/mo",
    "trafficNum": 720000,
    "searchVol": "120,000/mo",
    "searchVolNum": 120000,
    "searchTrend": "+165%",
    "latency": 14,
    "tokens": 96,
    "desc": "Array framework for machine learning research on Apple Silicon, engineered by Apple's machine learning research team.",
    "useCase": "Train and fine-tune large language models directly on Mac unified memory with NumPy-style Python APIs.",
    "arch": "C++, Metal shaders, Python unified memory arrays.",
    "tags": [
      "apple-silicon",
      "metal-acceleration",
      "unified-memory",
      "mac-ai"
    ],
    "elo": 1330,
    "eloCi": "±17",
    "winRate": 89,
    "starsDisplay": "28.7k",
    "weeklyStars": 10880,
    "weeklyGrowth": "+39.9%",
    "monthlyGrowth": "+57%",
    "wGrowthNum": 13.2,
    "mGrowthNum": 57,
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/ml-explore.png",
    "githubUrl": "https://github.com/ml-explore/mlx",
    "rank": 97
  },
  {
    "id": "axolotl",
    "name": "Axolotl",
    "repo": "axolotl-ai-cloud/axolotl",
    "author": "axolotl-ai-cloud",
    "persona": "Researchers & STEM",
    "targetRole": "Fine-Tuning Specialists",
    "stars": 12529,
    "wGrowth": 52.8,
    "mGrowth": 49.8,
    "website": "https://github.com/axolotl-ai-cloud/axolotl",
    "traffic": "410,000/mo",
    "trafficNum": 410000,
    "searchVol": "82,000/mo",
    "searchVolNum": 82000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Axolotl powering production AI workloads.",
    "useCase": "Integrate Axolotl into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1332,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "12.5k",
    "weeklyStars": 5403,
    "weeklyGrowth": "+52.8%",
    "monthlyGrowth": "+49.8%",
    "wGrowthNum": 13.1,
    "mGrowthNum": 49.8,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/axolotl-ai-cloud.png",
    "githubUrl": "https://github.com/axolotl-ai-cloud/axolotl",
    "rank": 98
  },
  {
    "id": "gemma-2",
    "name": "Google Gemma 2",
    "repo": "google-deepmind/gemma",
    "author": "google-deepmind",
    "persona": "Researchers & STEM",
    "targetRole": "Research Scientists",
    "stars": 5766,
    "wGrowth": 13,
    "mGrowth": 49.4,
    "website": "https://github.com/google-deepmind/gemma",
    "traffic": "970,000/mo",
    "trafficNum": 970000,
    "searchVol": "194,000/mo",
    "searchVolNum": 194000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Google Gemma 2 powering production AI workloads.",
    "useCase": "Integrate Google Gemma 2 into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1332,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "5.8k",
    "weeklyStars": 2522,
    "weeklyGrowth": "+13.0%",
    "monthlyGrowth": "+49.4%",
    "wGrowthNum": 13,
    "mGrowthNum": 49.4,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/google-deepmind.png",
    "githubUrl": "https://github.com/google-deepmind/gemma",
    "rank": 99
  },
  {
    "id": "open-webui",
    "name": "Open WebUI",
    "repo": "open-webui/open-webui",
    "author": "open-webui",
    "persona": "Founders & Solo Creators",
    "targetRole": "Self-Hosted AI Administrators",
    "stars": 154121,
    "wGrowth": 96.3,
    "mGrowth": 56,
    "website": "https://openwebui.com",
    "traffic": "2.8M/mo",
    "trafficNum": 2800000,
    "searchVol": "450,000/mo",
    "searchVolNum": 450000,
    "searchTrend": "+145%",
    "latency": 24,
    "tokens": 90,
    "desc": "Extensible, feature-rich, and user-friendly self-hosted WebUI for Ollama, OpenAI-compatible APIs, and local models.",
    "useCase": "Deploy a private multi-user ChatGPT interface for your team with RAG document chat, pipelines, and web search.",
    "arch": "Python FastAPI backend, SvelteKit frontend, Ollama/OpenAI connectors.",
    "tags": [
      "self-hosted-webui",
      "chatgpt-clone",
      "ollama-ui",
      "sveltekit"
    ],
    "elo": 1350,
    "eloCi": "±15",
    "winRate": 91,
    "starsDisplay": "154.1k",
    "weeklyStars": 85748,
    "weeklyGrowth": "+96.3%",
    "monthlyGrowth": "+56%",
    "wGrowthNum": 12.9,
    "mGrowthNum": 56,
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/open-webui.png",
    "githubUrl": "https://github.com/open-webui/open-webui",
    "rank": 100
  },
  {
    "id": "bolt-diy",
    "name": "Bolt.diy WebContainers",
    "repo": "stackblitz-labs/bolt.diy",
    "author": "stackblitz-labs",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Solo MVP Builders",
    "stars": 19933,
    "wGrowth": 60.8,
    "mGrowth": 54,
    "website": "https://stackblitz.com",
    "traffic": "3.2M/mo",
    "trafficNum": 3200000,
    "searchVol": "55,000/mo",
    "searchVolNum": 55000,
    "searchTrend": "+140%",
    "latency": 55,
    "tokens": 90,
    "desc": "In-browser Node.js sandbox that compiles, executes, and previews full-stack React and Next.js applications directly inside WebAssembly.",
    "useCase": "Prompt an entire SaaS landing page, backend API, and database schema into existence and preview it live without local setup.",
    "arch": "WebContainers WASM, Remix, Vite, WebSockets, LocalStorage runtime.",
    "tags": [
      "in-browser-build",
      "mvp-scaffold",
      "webcontainers",
      "zero-setup"
    ],
    "rank": 101,
    "elo": 1226,
    "eloCi": "±22",
    "winRate": 77,
    "starsDisplay": "19.9k",
    "weeklyStars": 9120,
    "weeklyGrowth": "+60.8%",
    "monthlyGrowth": "+54.0%",
    "wGrowthNum": 12.8,
    "mGrowthNum": 54,
    "latencyMs": "55ms",
    "latencyNum": 55,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/stackblitz-labs.png",
    "githubUrl": "https://github.com/stackblitz-labs/bolt.diy"
  },
  {
    "id": "whisper-turbo",
    "name": "Whisper Turbo",
    "repo": "openai/whisper-turbo",
    "author": "openai",
    "persona": "Music & Audio Creators",
    "targetRole": "Voice Developers",
    "stars": 16500,
    "wGrowth": 12.8,
    "mGrowth": 48.6,
    "website": "https://github.com/openai/whisper-turbo",
    "traffic": "825,000/mo",
    "trafficNum": 825000,
    "searchVol": "165,000/mo",
    "searchVolNum": 165000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Whisper Turbo powering production AI workloads.",
    "useCase": "Integrate Whisper Turbo into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1331,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "16.5k",
    "weeklyStars": 2112,
    "weeklyGrowth": "+12.8%",
    "monthlyGrowth": "+48.6%",
    "wGrowthNum": 12.8,
    "mGrowthNum": 48.6,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/openai.png",
    "githubUrl": "https://github.com/openai/whisper-turbo",
    "rank": 102
  },
  {
    "id": "trulens",
    "name": "TruLens",
    "repo": "truera/trulens",
    "author": "truera",
    "persona": "STEM & AI Researchers",
    "targetRole": "LLM Quality & Safety Auditors",
    "stars": 3592,
    "wGrowth": 12.8,
    "mGrowth": 54,
    "website": "https://trulens.org",
    "traffic": "160k/mo",
    "trafficNum": 160000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+145%",
    "latency": 22,
    "tokens": 90,
    "desc": "Evaluation and tracking tool to systematically measure and monitor LLM application quality with the RAG Triad.",
    "useCase": "Measure context relevance, groundedness, and answer relevance on every production query automatically.",
    "arch": "Python, Truera feedback functions, Streamlit dashboard.",
    "tags": [
      "rag-triad",
      "llm-benchmarking",
      "groundedness",
      "quality-metrics"
    ],
    "elo": 1265,
    "eloCi": "±21",
    "winRate": 82,
    "starsDisplay": "3.6k",
    "weeklyStars": 627,
    "weeklyGrowth": "+12.8%",
    "monthlyGrowth": "+54%",
    "wGrowthNum": 12.8,
    "mGrowthNum": 54,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/truera.png",
    "githubUrl": "https://github.com/truera/trulens",
    "rank": 103
  },
  {
    "id": "bitsandbytes",
    "name": "BitsAndBytes",
    "repo": "bitsandbytes-foundation/bitsandbytes",
    "author": "bitsandbytes-foundation",
    "persona": "STEM & AI Researchers",
    "targetRole": "Quantization Researchers",
    "stars": 8514,
    "wGrowth": 12.8,
    "mGrowth": 53,
    "website": "https://github.com/bitsandbytes-foundation/bitsandbytes",
    "traffic": "520k/mo",
    "searchVol": "88,000/mo",
    "latency": 10,
    "tokens": 99,
    "desc": "Accessible 8-bit and 4-bit CUDA kernels for matrix multiplication and quantization for LLMs.",
    "useCase": "Run 4-bit quantized base models with near 16-bit performance on low-VRAM gaming GPUs.",
    "arch": "CUDA, C++, PyTorch CGO bindings.",
    "tags": [
      "4-bit-quantization",
      "cuda-kernels",
      "gpu-memory",
      "nf4"
    ],
    "starsDisplay": "8.5k",
    "weeklyStars": 1712,
    "weeklyGrowth": "+12.8%",
    "monthlyGrowth": "+53%",
    "wGrowthNum": 12.8,
    "mGrowthNum": 53,
    "trafficNum": 520000,
    "searchVolNum": 88000,
    "searchTrend": "+128%",
    "latencyMs": "10ms",
    "latencyNum": 10,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "elo": 1310,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/bitsandbytes-foundation.png",
    "githubUrl": "https://github.com/bitsandbytes-foundation/bitsandbytes",
    "rank": 104
  },
  {
    "id": "candle",
    "name": "Hugging Face Candle",
    "repo": "huggingface/candle",
    "author": "huggingface",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Minimalist ML Systems Devs",
    "stars": 21143,
    "wGrowth": 25.9,
    "mGrowth": 55,
    "website": "https://github.com/huggingface/candle",
    "traffic": "780k/mo",
    "searchVol": "135,000/mo",
    "latency": 9,
    "tokens": 99,
    "desc": "Minimalist ML framework for Rust with focus on performance, ease of use, and serverless WASM inference.",
    "useCase": "Embed state-of-the-art transformer models directly into Rust edge binaries and browser WASM workers.",
    "arch": "Rust, CUDA, Metal, WebAssembly, Zero-PyTorch.",
    "tags": [
      "rust-ml",
      "wasm-ai",
      "minimalist-framework",
      "candle"
    ],
    "starsDisplay": "21.1k",
    "weeklyStars": 6493,
    "weeklyGrowth": "+25.9%",
    "monthlyGrowth": "+55%",
    "wGrowthNum": 12.8,
    "mGrowthNum": 55,
    "trafficNum": 780000,
    "searchVolNum": 135000,
    "searchTrend": "+128%",
    "latencyMs": "9ms",
    "latencyNum": 9,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "elo": 1311,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/candle",
    "rank": 105
  },
  {
    "id": "fluvio",
    "name": "Fluvio",
    "repo": "infinyon/fluvio",
    "author": "infinyon",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Real-Time Event Stream Leads",
    "stars": 5260,
    "starsDisplay": "5.3k",
    "weeklyStars": 2926,
    "weeklyGrowth": "+12.7%",
    "monthlyGrowth": "+48%",
    "wGrowthNum": 12.7,
    "mGrowthNum": 48,
    "traffic": "697k/mo",
    "trafficNum": 697435,
    "searchVol": "130,000/mo",
    "searchVolNum": 129894,
    "searchTrend": "+127%",
    "latencyMs": "11ms",
    "latencyNum": 17,
    "tokensSaved": "-97%",
    "tokensNum": 95,
    "elo": 1311,
    "eloCi": "±18",
    "winRate": 87,
    "website": "https://github.com/infinyon/fluvio",
    "githubUrl": "https://github.com/infinyon/fluvio",
    "logoUrl": "https://github.com/infinyon.png",
    "desc": "High-impact open-source repository empowering real-time event stream leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale streaming infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "streaming",
      "rust-broker",
      "edge-events",
      "fluvio"
    ],
    "rank": 106
  },
  {
    "id": "langgraph",
    "name": "LangGraph",
    "repo": "langchain-ai/langgraph",
    "author": "langchain-ai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "AI Systems Architects",
    "stars": 42806,
    "wGrowth": 99.1,
    "mGrowth": 48,
    "website": "https://langchain-ai.github.io/langgraph",
    "traffic": "3.2M/mo",
    "trafficNum": 3200000,
    "searchVol": "240,000/mo",
    "searchVolNum": 240000,
    "searchTrend": "+175%",
    "latency": 35,
    "tokens": 82,
    "desc": "Library for building stateful, multi-actor applications with LLMs, modeled as cyclical computation graphs.",
    "useCase": "Coordinate complex multi-agent swarms with human-in-the-loop approvals, time-travel debugging, and persistent state.",
    "arch": "Python / TypeScript, Pregel graph execution, SQLite / Postgres checkpointing.",
    "tags": [
      "graph-agent",
      "state-machine",
      "multi-agent",
      "langchain"
    ],
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/langchain-ai.png",
    "githubUrl": "https://github.com/langchain-ai/langgraph",
    "starsDisplay": "42.8k",
    "weeklyStars": 24015,
    "weeklyGrowth": "+99.1%",
    "monthlyGrowth": "+48.0%",
    "wGrowthNum": 12.6,
    "mGrowthNum": 48,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-82%",
    "tokensNum": 82,
    "rank": 107
  },
  {
    "id": "exllamav2",
    "name": "ExLlamaV2",
    "repo": "turboderp/exllamav2",
    "author": "turboderp",
    "persona": "Researchers & STEM",
    "targetRole": "Inference Engineers",
    "stars": 4630,
    "wGrowth": 12.5,
    "mGrowth": 47.5,
    "website": "https://github.com/turboderp/exllamav2",
    "traffic": "310,000/mo",
    "trafficNum": 310000,
    "searchVol": "62,000/mo",
    "searchVolNum": 62000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool ExLlamaV2 powering production AI workloads.",
    "useCase": "Integrate ExLlamaV2 into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1330,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "4.6k",
    "weeklyStars": 775,
    "weeklyGrowth": "+12.5%",
    "monthlyGrowth": "+47.5%",
    "wGrowthNum": 12.5,
    "mGrowthNum": 47.5,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/turboderp.png",
    "githubUrl": "https://github.com/turboderp/exllamav2",
    "rank": 108
  },
  {
    "id": "cloudflare-vectorize",
    "name": "Cloudflare Vectorize",
    "repo": "cloudflare/vectorize-examples",
    "author": "cloudflare",
    "persona": "AI Ops & Tooling",
    "targetRole": "Edge Engineers",
    "stars": 5200,
    "wGrowth": 12.4,
    "mGrowth": 47.1,
    "website": "https://github.com/cloudflare/vectorize-examples",
    "traffic": "260,000/mo",
    "trafficNum": 260000,
    "searchVol": "52,000/mo",
    "searchVolNum": 52000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Cloudflare Vectorize powering production AI workloads.",
    "useCase": "Integrate Cloudflare Vectorize into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1330,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "5.2k",
    "weeklyStars": 645,
    "weeklyGrowth": "+12.4%",
    "monthlyGrowth": "+47.1%",
    "wGrowthNum": 12.4,
    "mGrowthNum": 47.1,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/vectorize-examples",
    "rank": 109
  },
  {
    "id": "unstructured",
    "name": "Unstructured",
    "repo": "Unstructured-IO/unstructured",
    "author": "Unstructured-IO",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Data Ingestion Pipeline Leads",
    "stars": 15535,
    "wGrowth": 43.8,
    "mGrowth": 53,
    "website": "https://unstructured.io",
    "traffic": "890k/mo",
    "searchVol": "150,000/mo",
    "latency": 22,
    "tokens": 92,
    "desc": "Open-source components for ingesting and preprocessing unstructured data like PDF, HTML, Word, and images for RAG.",
    "useCase": "Clean and partition raw corporate documents into standardized document elements for downstream vector indexing.",
    "arch": "Python, Detectron2, Tesseract OCR, Partitioning algorithms.",
    "tags": [
      "data-preprocessing",
      "unstructured-data",
      "rag-etl",
      "document-ai"
    ],
    "starsDisplay": "15.5k",
    "weeklyStars": 6074,
    "weeklyGrowth": "+43.8%",
    "monthlyGrowth": "+53%",
    "wGrowthNum": 12.4,
    "mGrowthNum": 53,
    "trafficNum": 890000,
    "searchVolNum": 150000,
    "searchTrend": "+124%",
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "elo": 1307,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/Unstructured-IO.png",
    "githubUrl": "https://github.com/Unstructured-IO/unstructured",
    "rank": 110
  },
  {
    "id": "chainlit",
    "name": "Chainlit",
    "repo": "Chainlit/chainlit",
    "author": "Chainlit",
    "persona": "Founders & Solo Creators",
    "targetRole": "Conversational UI Developers",
    "stars": 12503,
    "wGrowth": 31.6,
    "mGrowth": 51,
    "website": "https://chainlit.io",
    "traffic": "510k/mo",
    "trafficNum": 510000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+140%",
    "latency": 22,
    "tokens": 86,
    "desc": "Build production-ready Conversational AI web interfaces and copilots in minutes with native multi-modal support.",
    "useCase": "Ship interactive chat apps with step visualizers, file uploads, and audio streaming using pure Python.",
    "arch": "Python backend, React frontend, WebSocket transport, OAuth2.",
    "tags": [
      "conversational-ui",
      "python-ui",
      "copilot-interface",
      "websockets"
    ],
    "elo": 1280,
    "eloCi": "±18",
    "winRate": 84,
    "starsDisplay": "12.5k",
    "weeklyStars": 4172,
    "weeklyGrowth": "+31.6%",
    "monthlyGrowth": "+51%",
    "wGrowthNum": 12.3,
    "mGrowthNum": 51,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "logoUrl": "https://github.com/Chainlit.png",
    "githubUrl": "https://github.com/Chainlit/chainlit",
    "rank": 111
  },
  {
    "id": "opencode-agent",
    "name": "OpenCode Interpreter",
    "repo": "OpenCode-AI/opencode",
    "author": "OpenCode-AI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Open Execution Sandboxes",
    "stars": 13789,
    "starsDisplay": "13.8k",
    "weeklyStars": 4261,
    "weeklyGrowth": "+12.3%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 12.3,
    "mGrowthNum": 46,
    "traffic": "520k/mo",
    "trafficNum": 519966,
    "searchVol": "75,000/mo",
    "searchVolNum": 74900,
    "searchTrend": "+123%",
    "latencyMs": "15ms",
    "latencyNum": 18,
    "tokensSaved": "-97%",
    "tokensNum": 93,
    "elo": 1309,
    "eloCi": "±18",
    "winRate": 87,
    "website": "https://github.com/OpenCode-AI/opencode",
    "githubUrl": "https://github.com/OpenCode-AI/opencode",
    "logoUrl": "https://github.com/OpenCode-AI.png",
    "desc": "High-impact open-source repository empowering open execution sandboxes with production capabilities.",
    "useCase": "Used by modern agentic teams to scale execution-sandbox infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "execution-sandbox",
      "python-eval",
      "security-jail",
      "agent-tools"
    ],
    "rank": 112
  },
  {
    "id": "garry-tan-playbook",
    "name": "Garry Tan Founder Playbook",
    "repo": "garrytan/founder-playbook",
    "author": "garrytan",
    "persona": "Founders & Indie Hackers",
    "targetRole": "YC Founders & Early-Stage CEOs",
    "stars": 18200,
    "wGrowth": "+12.1%",
    "mGrowth": "+52.3%",
    "website": "https://garrytan.com",
    "traffic": "520k/mo",
    "trafficNum": 520000,
    "searchVol": "68,000/mo",
    "searchVolNum": 68000,
    "searchTrend": "+88% (Surging)",
    "latency": "15ms",
    "tokens": "-60%",
    "desc": "Curated knowledge base and agentic reasoning framework based on Y Combinator CEO Garry Tan's founder playbooks, pitch diagnostics, and PMF heuristics.",
    "useCase": "Advising early-stage founders through automated YC application evaluations, investor deck stress tests, and pricing strategy reviews.",
    "arch": "Markdown knowledge graphs, Vector RAG index, Jev scoring heuristics.",
    "tags": [
      "Garry Tan",
      "Y Combinator",
      "Founders",
      "Startup PMF",
      "Venture Capital"
    ],
    "wGrowthNum": 12.1,
    "mGrowthNum": 52.3,
    "latencyNum": 15,
    "tokensNum": 60,
    "weeklyStars": 2202,
    "weeklyGrowth": "+12.1%",
    "monthlyGrowth": "+52.3%",
    "latencyMs": "15ms",
    "tokensSaved": "-60%",
    "starsDisplay": "18.2k",
    "logoUrl": "https://github.com/garrytan.png",
    "githubUrl": "https://github.com/garrytan/founder-playbook",
    "rank": 113,
    "elo": 1232,
    "eloCi": "±23",
    "winRate": 78
  },
  {
    "id": "windmill",
    "name": "Windmill",
    "repo": "windmill-labs/windmill",
    "author": "windmill-labs",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Internal Tools & Platform Engineers",
    "stars": 18119,
    "wGrowth": 25,
    "mGrowth": 49,
    "website": "https://windmill.dev",
    "traffic": "560k/mo",
    "trafficNum": 560000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+170%",
    "latency": 15,
    "tokens": 94,
    "desc": "Open-source developer platform to turn scripts into internal tools, workflows, and UIs with Rust performance.",
    "useCase": "Replace rigid Retool apps with fast, version-controlled scripts executed deterministically on distributed workers.",
    "arch": "Rust core, Svelte frontend, PostgreSQL, Docker worker daemon.",
    "tags": [
      "internal-tools",
      "workflow-engine",
      "rust",
      "low-code"
    ],
    "elo": 1310,
    "eloCi": "±19",
    "winRate": 86,
    "logoUrl": "https://github.com/windmill-labs.png",
    "githubUrl": "https://github.com/windmill-labs/windmill",
    "starsDisplay": "18.1k",
    "weeklyStars": 5374,
    "weeklyGrowth": "+25%",
    "monthlyGrowth": "+49.0%",
    "wGrowthNum": 12.1,
    "mGrowthNum": 49,
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "rank": 114
  },
  {
    "id": "cloudflare-queues",
    "name": "Cloudflare Queues",
    "repo": "cloudflare/queues-worker-demo",
    "author": "cloudflare",
    "persona": "AI Ops & Tooling",
    "targetRole": "Distributed Systems Ops",
    "stars": 3900,
    "wGrowth": 12.1,
    "mGrowth": 46,
    "website": "https://github.com/cloudflare/queues-worker-demo",
    "traffic": "195,000/mo",
    "trafficNum": 195000,
    "searchVol": "39,000/mo",
    "searchVolNum": 39000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Cloudflare Queues powering production AI workloads.",
    "useCase": "Integrate Cloudflare Queues into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1328,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "3.9k",
    "weeklyStars": 472,
    "weeklyGrowth": "+12.1%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 12.1,
    "mGrowthNum": 46,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/queues-worker-demo",
    "rank": 115
  },
  {
    "id": "txtai",
    "name": "txtai",
    "repo": "neuml/txtai",
    "author": "neuml",
    "persona": "Autonomous Software Engineers",
    "targetRole": "All-in-One AI Search Leads",
    "stars": 12992,
    "wGrowth": 24.9,
    "mGrowth": 51,
    "website": "https://neuml.github.io/txtai",
    "traffic": "410k/mo",
    "searchVol": "72,000/mo",
    "latency": 16,
    "tokens": 95,
    "desc": "All-in-one embeddings database for semantic search, LLM orchestration, and language model workflows.",
    "useCase": "Run semantic vector search, relational queries, and automated audio/text transcription inside a single lightweight framework.",
    "arch": "Python, Faiss, PyTorch, Transformers, SQLite.",
    "tags": [
      "semantic-search",
      "embeddings-database",
      "rag-pipeline",
      "neuml"
    ],
    "starsDisplay": "13.0k",
    "weeklyStars": 3850,
    "weeklyGrowth": "+24.9%",
    "monthlyGrowth": "+51%",
    "wGrowthNum": 12.1,
    "mGrowthNum": 51,
    "trafficNum": 410000,
    "searchVolNum": 72000,
    "searchTrend": "+121%",
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "elo": 1305,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/neuml.png",
    "githubUrl": "https://github.com/neuml/txtai",
    "rank": 116
  },
  {
    "id": "neovim",
    "name": "Neovim",
    "repo": "neovim/neovim",
    "author": "neovim",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Extensible Terminal Editor Leads",
    "stars": 102889,
    "starsDisplay": "102.9k",
    "weeklyStars": 90215,
    "weeklyGrowth": "+613.6%",
    "monthlyGrowth": "+45%",
    "wGrowthNum": 12.1,
    "mGrowthNum": 45,
    "traffic": "565k/mo",
    "trafficNum": 565126,
    "searchVol": "62,000/mo",
    "searchVolNum": 61789,
    "searchTrend": "+121%",
    "latencyMs": "25ms",
    "latencyNum": 21,
    "tokensSaved": "-96%",
    "tokensNum": 95,
    "elo": 1306,
    "eloCi": "±18",
    "winRate": 87,
    "website": "https://github.com/neovim/neovim",
    "githubUrl": "https://github.com/neovim/neovim",
    "logoUrl": "https://github.com/neovim.png",
    "desc": "High-impact open-source tool empowering extensible terminal editor leads with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical lua-runtime infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "lua-runtime",
      "lsp-client",
      "modal-editing",
      "neovim"
    ],
    "rank": 117,
    "wGrowth": 613.6
  },
  {
    "id": "apple-mlx",
    "name": "Apple MLX",
    "repo": "ml-explore/mlx",
    "author": "ml-explore",
    "persona": "Researchers & STEM",
    "targetRole": "Apple Silicon AI Researchers",
    "stars": 28674,
    "wGrowth": 44.8,
    "mGrowth": 47,
    "website": "https://ml-explore.github.io/mlx",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "190,000/mo",
    "searchVolNum": 190000,
    "searchTrend": "+175%",
    "latency": 18,
    "tokens": 94,
    "desc": "Array framework for machine learning research on Apple silicon, designed by Apple's machine learning research team.",
    "useCase": "Train and run large language models utilizing unified memory on Mac Studio and MacBook Pro hardware at peak efficiency.",
    "arch": "C++, Python API, Metal unified memory, Lazy graph evaluation.",
    "tags": [
      "apple-silicon",
      "metal",
      "machine-learning",
      "unified-memory"
    ],
    "elo": 1330,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/ml-explore.png",
    "githubUrl": "https://github.com/ml-explore/mlx",
    "starsDisplay": "28.7k",
    "weeklyStars": 11250,
    "weeklyGrowth": "+44.8%",
    "monthlyGrowth": "+47.0%",
    "wGrowthNum": 12,
    "mGrowthNum": 47,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "rank": 118
  },
  {
    "id": "trigger-dev",
    "name": "Trigger.dev",
    "repo": "triggerdotdev/trigger.dev",
    "author": "triggerdotdev",
    "persona": "Founders & Solo Creators",
    "targetRole": "Background Job Leads",
    "stars": 16490,
    "starsDisplay": "16.5k",
    "weeklyStars": 4282,
    "weeklyGrowth": "+12.0%",
    "monthlyGrowth": "+45%",
    "wGrowthNum": 12,
    "mGrowthNum": 45,
    "traffic": "577k/mo",
    "trafficNum": 577321,
    "searchVol": "69,000/mo",
    "searchVolNum": 68997,
    "searchTrend": "+120%",
    "latencyMs": "15ms",
    "latencyNum": 11,
    "tokensSaved": "-89%",
    "tokensNum": 88,
    "elo": 1307,
    "eloCi": "±18",
    "winRate": 87,
    "website": "https://github.com/triggerdotdev/trigger.dev",
    "githubUrl": "https://github.com/triggerdotdev/trigger.dev",
    "logoUrl": "https://github.com/triggerdotdev.png",
    "desc": "High-impact open-source repository empowering background job leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale background-jobs infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "background-jobs",
      "long-running-tasks",
      "realtime",
      "nextjs"
    ],
    "rank": 119
  },
  {
    "id": "cloudflare-hyperdrive",
    "name": "Cloudflare Hyperdrive",
    "repo": "cloudflare/hyperdrive-starter",
    "author": "cloudflare",
    "persona": "AI Ops & Tooling",
    "targetRole": "Database Engineers",
    "stars": 4100,
    "wGrowth": 11.9,
    "mGrowth": 45.2,
    "website": "https://github.com/cloudflare/hyperdrive-starter",
    "traffic": "205,000/mo",
    "trafficNum": 205000,
    "searchVol": "41,000/mo",
    "searchVolNum": 41000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Cloudflare Hyperdrive powering production AI workloads.",
    "useCase": "Integrate Cloudflare Hyperdrive into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1328,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "4.1k",
    "weeklyStars": 488,
    "weeklyGrowth": "+11.9%",
    "monthlyGrowth": "+45.2%",
    "wGrowthNum": 11.9,
    "mGrowthNum": 45.2,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/hyperdrive-starter",
    "rank": 120
  },
  {
    "id": "crawl4ai",
    "name": "Crawl4AI",
    "repo": "unclecode/crawl4ai",
    "author": "unclecode",
    "persona": "AI Ops & Tooling",
    "targetRole": "Data Extraction & Agent Devs",
    "stars": 84873,
    "wGrowth": 148.2,
    "mGrowth": 51,
    "website": "https://crawl4ai.com",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "85,000/mo",
    "searchVolNum": 85000,
    "searchTrend": "+120%",
    "latency": 38,
    "tokens": 96,
    "desc": "Open-source asynchronous LLM-friendly web crawler & scraper that outputs clean, pristine markdown with smart chunking.",
    "useCase": "Scrape heavy JS-rendered documentation, blog posts, and dynamic tables into clean markdown for agent context windows.",
    "arch": "Playwright headless, asyncio, BeautifulSoup4, LXML, Python.",
    "tags": [
      "scraping",
      "llm-markdown",
      "async-crawler",
      "headless"
    ],
    "rank": 121,
    "elo": 1262,
    "eloCi": "±24",
    "winRate": 82,
    "starsDisplay": "84.9k",
    "weeklyStars": 54708,
    "weeklyGrowth": "+148.2%",
    "monthlyGrowth": "+51.0%",
    "wGrowthNum": 11.8,
    "mGrowthNum": 51,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/unclecode.png",
    "githubUrl": "https://github.com/unclecode/crawl4ai"
  },
  {
    "id": "cline-code",
    "name": "Cline",
    "repo": "cline/cline",
    "author": "cline",
    "persona": "Autonomous Software Engineers",
    "targetRole": "VSCode Power Users",
    "stars": 69962,
    "wGrowth": 185.6,
    "mGrowth": 51,
    "website": "https://cline.bot",
    "traffic": "1.6M/mo",
    "trafficNum": 1600000,
    "searchVol": "85,000/mo",
    "searchVolNum": 85000,
    "searchTrend": "+125%",
    "latency": 50,
    "tokens": 93,
    "desc": "Autonomous coding agent in VS Code that inspects project files, runs terminal commands, and edits code with Model Context Protocol support.",
    "useCase": "Hand off complex tasks like 'upgrade Next.js 14 to 15 and fix all deprecated breaking changes' and watch it complete and verify.",
    "arch": "VSCode Extension API, Model Context Protocol, Diff Match Patch.",
    "tags": [
      "vscode-agent",
      "mcp-tools",
      "autonomous-coding",
      "diffs"
    ],
    "rank": 122,
    "elo": 1243,
    "eloCi": "±25",
    "winRate": 79,
    "starsDisplay": "70.0k",
    "weeklyStars": 48353,
    "weeklyGrowth": "+185.6%",
    "monthlyGrowth": "+51.0%",
    "wGrowthNum": 11.8,
    "mGrowthNum": 51,
    "latencyMs": "50ms",
    "latencyNum": 50,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/cline.png",
    "githubUrl": "https://github.com/cline/cline"
  },
  {
    "id": "dify",
    "name": "Dify.AI",
    "repo": "langgenius/dify",
    "author": "langgenius",
    "persona": "Founders & Indie Hackers",
    "targetRole": "AI Product Founders",
    "stars": 157988,
    "wGrowth": 146.1,
    "mGrowth": 46,
    "website": "https://dify.ai",
    "traffic": "4.8M/mo",
    "trafficNum": 4800000,
    "searchVol": "520,000/mo",
    "searchVolNum": 520000,
    "searchTrend": "+185%",
    "latency": 32,
    "tokens": 86,
    "desc": "Open-source LLM app development platform combining AI workflow, RAG pipeline, agent capabilities, and model management.",
    "useCase": "Visually build, test, and ship complete conversational AI applications with enterprise-grade RAG and API access.",
    "arch": "Python / Flask, Next.js, Celery, Redis, PostgreSQL, Weaviate.",
    "tags": [
      "llm-platform",
      "visual-workflow",
      "rag-engine",
      "enterprise-agent"
    ],
    "elo": 1355,
    "eloCi": "±16",
    "winRate": 92,
    "logoUrl": "https://github.com/langgenius.png",
    "githubUrl": "https://github.com/langgenius/dify",
    "starsDisplay": "158.0k",
    "weeklyStars": 101364,
    "weeklyGrowth": "+146.1%",
    "monthlyGrowth": "+46.0%",
    "wGrowthNum": 11.8,
    "mGrowthNum": 46,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "rank": 123
  },
  {
    "id": "sqlglot",
    "name": "SQLGlot",
    "repo": "tobymao/sqlglot",
    "author": "tobymao",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Data Infrastructure Engineers",
    "stars": 9664,
    "wGrowth": 15,
    "mGrowth": 48,
    "website": "https://sqlglot.com",
    "traffic": "410k/mo",
    "trafficNum": 410000,
    "searchVol": "68,000/mo",
    "searchVolNum": 68000,
    "searchTrend": "+125%",
    "latency": 12,
    "tokens": 98,
    "desc": "An ultra-fast, zero-dependency Python SQL parser, transpiler, optimizer, and engine supporting 21 dialect translations.",
    "useCase": "Transpile complex queries between BigQuery, Snowflake, DuckDB, and Postgres with guaranteed semantic parity.",
    "arch": "Pure Python, AST parser, Dialect generators, Optimizer rules.",
    "tags": [
      "sql-parser",
      "transpiler",
      "zero-dependency",
      "data-infra"
    ],
    "elo": 1285,
    "eloCi": "±19",
    "winRate": 84,
    "starsDisplay": "9.7k",
    "weeklyStars": 2255,
    "weeklyGrowth": "+15%",
    "monthlyGrowth": "+48%",
    "wGrowthNum": 11.8,
    "mGrowthNum": 48,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/tobymao.png",
    "githubUrl": "https://github.com/tobymao/sqlglot",
    "rank": 124
  },
  {
    "id": "comfyui",
    "name": "ComfyUI",
    "repo": "comfyanonymous/ComfyUI",
    "author": "comfyanonymous",
    "persona": "Music & Audio Creators",
    "targetRole": "Visual Directors & Generative Artists",
    "stars": 136375,
    "wGrowth": 118.5,
    "mGrowth": 46,
    "website": "https://comfy.org",
    "traffic": "5.4M/mo",
    "trafficNum": 5400000,
    "searchVol": "680,000/mo",
    "searchVolNum": 680000,
    "searchTrend": "+175%",
    "latency": 48,
    "tokens": 88,
    "desc": "Most powerful and modular node-based visual interface and backend for Stable Diffusion, Flux, and SDXL.",
    "useCase": "Design reproducible, node-based generative pipelines combining image-to-video, inpainting, and LoRA chaining.",
    "arch": "Python, PyTorch, LiteGraph.js, Async execution graph.",
    "tags": [
      "stable-diffusion",
      "visual-nodes",
      "generative-art",
      "flux"
    ],
    "elo": 1360,
    "eloCi": "±16",
    "winRate": 92,
    "logoUrl": "https://github.com/comfyanonymous.png",
    "githubUrl": "https://github.com/comfyanonymous/ComfyUI",
    "starsDisplay": "136.4k",
    "weeklyStars": 81213,
    "weeklyGrowth": "+118.5%",
    "monthlyGrowth": "+46.0%",
    "wGrowthNum": 11.6,
    "mGrowthNum": 46,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 125
  },
  {
    "id": "langflow",
    "name": "Langflow",
    "repo": "langflow-ai/langflow",
    "author": "langflow-ai",
    "persona": "Founders & Indie Hackers",
    "targetRole": "AI Solutions Architects",
    "stars": 155554,
    "wGrowth": 222.7,
    "mGrowth": 45,
    "website": "https://langflow.org",
    "traffic": "2.3M/mo",
    "trafficNum": 2300000,
    "searchVol": "340,000/mo",
    "searchVolNum": 340000,
    "searchTrend": "+160%",
    "latency": 30,
    "tokens": 87,
    "desc": "Dynamic graph-based agent builder and visual prototyping environment for Python AI applications.",
    "useCase": "Assemble complex multi-agent architectures visually and export directly to production Python execution code.",
    "arch": "Python, FastAPI, React, Astra DB, LangChain integration.",
    "tags": [
      "visual-builder",
      "graph-orchestration",
      "fastapi",
      "prototyping"
    ],
    "elo": 1330,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/langflow-ai.png",
    "githubUrl": "https://github.com/langflow-ai/langflow",
    "starsDisplay": "155.6k",
    "weeklyStars": 112897,
    "weeklyGrowth": "+222.7%",
    "monthlyGrowth": "+45.0%",
    "wGrowthNum": 11.5,
    "mGrowthNum": 45,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-87%",
    "tokensNum": 87,
    "rank": 126
  },
  {
    "id": "tensorrt-llm",
    "name": "TensorRT-LLM",
    "repo": "NVIDIA/TensorRT-LLM",
    "author": "NVIDIA",
    "persona": "Researchers & STEM",
    "targetRole": "GPU Optimization Engineers",
    "stars": 14772,
    "wGrowth": 36.8,
    "mGrowth": 43.7,
    "website": "https://github.com/NVIDIA/TensorRT-LLM",
    "traffic": "540,000/mo",
    "trafficNum": 540000,
    "searchVol": "108,000/mo",
    "searchVolNum": 108000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool TensorRT-LLM powering production AI workloads.",
    "useCase": "Integrate TensorRT-LLM into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1326,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "14.8k",
    "weeklyStars": 5214,
    "weeklyGrowth": "+36.8%",
    "monthlyGrowth": "+43.7%",
    "wGrowthNum": 11.5,
    "mGrowthNum": 43.7,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/NVIDIA.png",
    "githubUrl": "https://github.com/NVIDIA/TensorRT-LLM",
    "rank": 127
  },
  {
    "id": "localai",
    "name": "LocalAI",
    "repo": "mudler/LocalAI",
    "author": "mudler",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Self-Hosted Cloud Architects",
    "stars": 49417,
    "wGrowth": 73.4,
    "mGrowth": 49,
    "website": "https://localai.io",
    "traffic": "460k/mo",
    "trafficNum": 460000,
    "searchVol": "75,000/mo",
    "searchVolNum": 75000,
    "searchTrend": "+130%",
    "latency": 28,
    "tokens": 91,
    "desc": "Free, open-source drop-in replacement for OpenAI API running locally on CPU or consumer GPU with zero internet access.",
    "useCase": "Self-host an OpenAI-compatible REST endpoint on internal VPS clusters for embeddings, audio, and vision.",
    "arch": "Go, gRPC backends, CGO bindings, Dockerized microservices.",
    "tags": [
      "openai-compatible",
      "self-hosted",
      "local-inference",
      "go-backend"
    ],
    "elo": 1295,
    "eloCi": "±18",
    "winRate": 85,
    "starsDisplay": "49.4k",
    "weeklyStars": 24195,
    "weeklyGrowth": "+73.4%",
    "monthlyGrowth": "+49%",
    "wGrowthNum": 11.5,
    "mGrowthNum": 49,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/mudler.png",
    "githubUrl": "https://github.com/mudler/LocalAI",
    "rank": 128
  },
  {
    "id": "safetensors",
    "name": "Safetensors",
    "repo": "huggingface/safetensors",
    "author": "huggingface",
    "persona": "STEM & AI Researchers",
    "targetRole": "Model Serialization Specialists",
    "stars": 3910,
    "wGrowth": 11.5,
    "mGrowth": 49,
    "website": "https://github.com/huggingface/safetensors",
    "traffic": "890k/mo",
    "searchVol": "155,000/mo",
    "latency": 6,
    "tokens": 100,
    "desc": "Simple, safe new format for storing tensors quickly and securely without arbitrary code execution risk.",
    "useCase": "Load multi-gigabyte neural network weights via zero-copy memory mapping without security risks of pickle.",
    "arch": "Rust, C bindings, Python wrappers, Memory-mapped IO.",
    "tags": [
      "zero-copy-load",
      "tensor-serialization",
      "safety-format",
      "huggingface"
    ],
    "starsDisplay": "3.9k",
    "weeklyStars": 459,
    "weeklyGrowth": "+11.5%",
    "monthlyGrowth": "+49%",
    "wGrowthNum": 11.5,
    "mGrowthNum": 49,
    "trafficNum": 890000,
    "searchVolNum": 155000,
    "searchTrend": "+115%",
    "latencyMs": "6ms",
    "latencyNum": 6,
    "tokensSaved": "-100%",
    "tokensNum": 100,
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/safetensors",
    "rank": 129
  },
  {
    "id": "clickhouse",
    "name": "ClickHouse",
    "repo": "ClickHouse/ClickHouse",
    "author": "ClickHouse",
    "persona": "STEM & AI Researchers",
    "targetRole": "Real-Time Analytics Architects",
    "stars": 50278,
    "starsDisplay": "50.3k",
    "weeklyStars": 20250,
    "weeklyGrowth": "+48.2%",
    "monthlyGrowth": "+43%",
    "wGrowthNum": 11.5,
    "mGrowthNum": 43,
    "traffic": "317k/mo",
    "trafficNum": 316787,
    "searchVol": "83,000/mo",
    "searchVolNum": 83332,
    "searchTrend": "+115%",
    "latencyMs": "22ms",
    "latencyNum": 27,
    "tokensSaved": "-92%",
    "tokensNum": 89,
    "elo": 1303,
    "eloCi": "±18",
    "winRate": 86,
    "website": "https://github.com/ClickHouse/ClickHouse",
    "githubUrl": "https://github.com/ClickHouse/ClickHouse",
    "logoUrl": "https://github.com/ClickHouse.png",
    "desc": "High-impact open-source repository empowering real-time analytics architects with production capabilities.",
    "useCase": "Used by modern agentic teams to scale columnar-db infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "columnar-db",
      "olap",
      "billion-scale",
      "realtime-analytics"
    ],
    "rank": 130,
    "wGrowth": 48.2
  },
  {
    "id": "cloudflare-agents-sdk",
    "name": "Cloudflare Agents SDK",
    "repo": "cloudflare/agents-sdk",
    "author": "cloudflare",
    "persona": "AI Ops & Tooling",
    "targetRole": "AI Platform Engineers & Edge Builders",
    "stars": 12400,
    "wGrowth": "+11.4%",
    "mGrowth": "+48.2%",
    "website": "https://developers.cloudflare.com/agents",
    "traffic": "420k/mo",
    "trafficNum": 420000,
    "searchVol": "38,000/mo",
    "searchVolNum": 38000,
    "searchTrend": "+74% (Surging)",
    "latency": "28ms",
    "tokens": "-45%",
    "desc": "Stateful, autonomous AI agents deployed across 330+ Cloudflare edge locations with persistent memory via Durable Objects and Vectorize.",
    "useCase": "Deploying real-time multi-tenant autonomous agents with sub-30ms global edge latency and zero server management.",
    "arch": "V8 Isolates, Cloudflare Workers, Durable Objects, Vectorize, WebSockets.",
    "tags": [
      "Edge AI",
      "Durable Objects",
      "Serverless",
      "Stateful Agents",
      "Cloudflare"
    ],
    "wGrowthNum": 11.4,
    "mGrowthNum": 48.2,
    "latencyNum": 28,
    "tokensNum": 45,
    "weeklyStars": 1413,
    "weeklyGrowth": "+11.4%",
    "monthlyGrowth": "+48.2%",
    "latencyMs": "28ms",
    "tokensSaved": "-45%",
    "starsDisplay": "12.4k",
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/agents-sdk",
    "rank": 131,
    "elo": 1215,
    "eloCi": "±26",
    "winRate": 75
  },
  {
    "id": "crewai",
    "name": "CrewAI",
    "repo": "crewAIInc/crewAI",
    "author": "crewAIInc",
    "persona": "Autonomous Software Engineers",
    "targetRole": "AI Product Engineers",
    "stars": 59408,
    "wGrowth": 109.2,
    "mGrowth": 44,
    "website": "https://crewai.com",
    "traffic": "1.9M/mo",
    "trafficNum": 1900000,
    "searchVol": "280,000/mo",
    "searchVolNum": 280000,
    "searchTrend": "+150%",
    "latency": 38,
    "tokens": 80,
    "desc": "Framework for orchestrating role-playing, autonomous AI agents that foster collaborative intelligence.",
    "useCase": "Assign distinct specialist roles (Researcher, Writer, QA, Marketer) to collaborate on multi-step deliverables.",
    "arch": "Python, LangChain integration, Telemetry, Task pipelines.",
    "tags": [
      "role-playing-agents",
      "orchestration",
      "multi-agent",
      "automation"
    ],
    "elo": 1310,
    "eloCi": "±20",
    "winRate": 85,
    "logoUrl": "https://github.com/crewAIInc.png",
    "githubUrl": "https://github.com/crewAIInc/crewAI",
    "starsDisplay": "59.4k",
    "weeklyStars": 34246,
    "weeklyGrowth": "+109.2%",
    "monthlyGrowth": "+44.0%",
    "wGrowthNum": 11.4,
    "mGrowthNum": 44,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-80%",
    "tokensNum": 80,
    "rank": 132
  },
  {
    "id": "starship",
    "name": "Starship Prompt",
    "repo": "starship/starship",
    "author": "starship",
    "persona": "Founders & Solo Creators",
    "targetRole": "Cross-Shell CLI Designers",
    "stars": 60175,
    "starsDisplay": "60.2k",
    "weeklyStars": 31399,
    "weeklyGrowth": "+85.3%",
    "monthlyGrowth": "+43%",
    "wGrowthNum": 11.4,
    "mGrowthNum": 43,
    "traffic": "681k/mo",
    "trafficNum": 681108,
    "searchVol": "53,000/mo",
    "searchVolNum": 53381,
    "searchTrend": "+114%",
    "latencyMs": "23ms",
    "latencyNum": 19,
    "tokensSaved": "-90%",
    "tokensNum": 98,
    "elo": 1303,
    "eloCi": "±18",
    "winRate": 86,
    "website": "https://github.com/starship/starship",
    "githubUrl": "https://github.com/starship/starship",
    "logoUrl": "https://github.com/starship.png",
    "desc": "High-impact open-source tool empowering cross-shell cli designers with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical prompt infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "prompt",
      "rust-cli",
      "cross-shell",
      "developer-experience"
    ],
    "rank": 133,
    "wGrowth": 85.3
  },
  {
    "id": "claude-code",
    "name": "Claude Code CLI",
    "repo": "anthropics/claude-code",
    "author": "anthropics",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Terminal Coding Pioneers",
    "stars": 149675,
    "starsDisplay": "149.7k",
    "weeklyStars": 138545,
    "weeklyGrowth": "+1092.8%",
    "monthlyGrowth": "+42%",
    "wGrowthNum": 11.3,
    "mGrowthNum": 42,
    "traffic": "610k/mo",
    "trafficNum": 610490,
    "searchVol": "104,000/mo",
    "searchVolNum": 103930,
    "searchTrend": "+113%",
    "latencyMs": "28ms",
    "latencyNum": 26,
    "tokensSaved": "-91%",
    "tokensNum": 97,
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "website": "https://github.com/anthropics/claude-code",
    "githubUrl": "https://github.com/anthropics/claude-code",
    "logoUrl": "https://github.com/anthropics.png",
    "desc": "High-impact open-source repository empowering terminal coding pioneers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale claude-code infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "claude-code",
      "terminal-swe",
      "anthropic-cli",
      "agentic-coding"
    ],
    "rank": 134,
    "wGrowth": 1092.8
  },
  {
    "id": "fish-speech",
    "name": "Fish Speech",
    "repo": "fishaudio/fish-speech",
    "author": "fishaudio",
    "persona": "Music & Audio Creators",
    "targetRole": "Vocal Synthesists",
    "stars": 32955,
    "wGrowth": 96.2,
    "mGrowth": 48,
    "website": "https://fish.audio",
    "traffic": "620k/mo",
    "trafficNum": 620000,
    "searchVol": "54,000/mo",
    "searchVolNum": 54000,
    "searchTrend": "+115%",
    "latency": 58,
    "tokens": 93,
    "desc": "Open-source text-to-speech architecture using dual-autoregressive and VQ-GAN tokens.",
    "useCase": "Clone vocal timbres for character singing and narration.",
    "arch": "Dual-AR Transformer, VQ-GAN, FlashAttention.",
    "tags": [
      "voice-synthesis",
      "tts",
      "open-weights"
    ],
    "rank": 135,
    "elo": 1222,
    "eloCi": "±18",
    "winRate": 76,
    "starsDisplay": "33.0k",
    "weeklyStars": 18036,
    "weeklyGrowth": "+96.2%",
    "monthlyGrowth": "+48.0%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 48,
    "latencyMs": "58ms",
    "latencyNum": 58,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/fishaudio.png",
    "githubUrl": "https://github.com/fishaudio/fish-speech"
  },
  {
    "id": "plandex",
    "name": "Plandex",
    "repo": "plandex-ai/plandex",
    "author": "plandex-ai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Senior Software Engineers",
    "stars": 15701,
    "wGrowth": 11.2,
    "mGrowth": 46,
    "website": "https://plandex.ai",
    "traffic": "380k/mo",
    "trafficNum": 380000,
    "searchVol": "72,000/mo",
    "searchVolNum": 72000,
    "searchTrend": "+140%",
    "latency": 34,
    "tokens": 86,
    "desc": "AI-driven coding engine designed for complex, multi-file software engineering tasks with branches, diffs, and rollbacks.",
    "useCase": "Break down massive feature roadmaps into plan trees, review syntax diffs before applying, and commit atomically.",
    "arch": "Go, Tree-sitter, Git integration, SQLite state store.",
    "tags": [
      "coding-engine",
      "multi-file",
      "go",
      "version-control"
    ],
    "elo": 1290,
    "eloCi": "±21",
    "winRate": 84,
    "logoUrl": "https://github.com/plandex-ai.png",
    "githubUrl": "https://github.com/plandex-ai/plandex",
    "starsDisplay": "15.7k",
    "weeklyStars": 2026,
    "weeklyGrowth": "+11.2%",
    "monthlyGrowth": "+46.0%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "latencyMs": "34ms",
    "latencyNum": 34,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "rank": 136
  },
  {
    "id": "duckdb",
    "name": "DuckDB",
    "repo": "duckdb/duckdb",
    "author": "duckdb",
    "persona": "Researchers & STEM",
    "targetRole": "Analytics & Data Platform Engineers",
    "stars": 41956,
    "wGrowth": 45.2,
    "mGrowth": 44,
    "website": "https://duckdb.org",
    "traffic": "2.8M/mo",
    "trafficNum": 2800000,
    "searchVol": "410,000/mo",
    "searchVolNum": 410000,
    "searchTrend": "+165%",
    "latency": 12,
    "tokens": 96,
    "desc": "In-process SQL OLAP database management system. Fast analytical queries on Parquet, CSV, and JSON files without servers.",
    "useCase": "Run instant sub-second analytical aggregations over local Parquet dumps directly inside client Python/Wasm processes.",
    "arch": "C++, Columnar vectorized execution, Zero-copy memory access.",
    "tags": [
      "olap",
      "sql-database",
      "in-process",
      "parquet"
    ],
    "elo": 1355,
    "eloCi": "±16",
    "winRate": 92,
    "logoUrl": "https://github.com/duckdb.png",
    "githubUrl": "https://github.com/duckdb/duckdb",
    "starsDisplay": "42.0k",
    "weeklyStars": 16293,
    "weeklyGrowth": "+45.2%",
    "monthlyGrowth": "+44.0%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 44,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "rank": 137
  },
  {
    "id": "openvoice",
    "name": "OpenVoice",
    "repo": "myshell-ai/OpenVoice",
    "author": "myshell-ai",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Engineers",
    "stars": 37794,
    "wGrowth": 21.1,
    "mGrowth": 42.6,
    "website": "https://github.com/myshell-ai/OpenVoice",
    "traffic": "1,560,000/mo",
    "trafficNum": 1560000,
    "searchVol": "312,000/mo",
    "searchVolNum": 312000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool OpenVoice powering production AI workloads.",
    "useCase": "Integrate OpenVoice into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1325,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "37.8k",
    "weeklyStars": 10088,
    "weeklyGrowth": "+21.1%",
    "monthlyGrowth": "+42.6%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 42.6,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/myshell-ai.png",
    "githubUrl": "https://github.com/myshell-ai/OpenVoice",
    "rank": 138
  },
  {
    "id": "haystack",
    "name": "Haystack",
    "repo": "deepset-ai/haystack",
    "author": "deepset-ai",
    "persona": "STEM & AI Researchers",
    "targetRole": "Enterprise RAG Architects",
    "stars": 26685,
    "wGrowth": 55.1,
    "mGrowth": 46,
    "website": "https://haystack.deepset.ai",
    "traffic": "620k/mo",
    "trafficNum": 620000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+120%",
    "latency": 28,
    "tokens": 84,
    "desc": "End-to-end LLM orchestration framework for building customizable, production-grade search and RAG pipelines.",
    "useCase": "Assemble deterministic multi-component agent pipelines with granular telemetry and vector database connectors.",
    "arch": "Python, Async pipelines, OpenTelemetry, Pydantic.",
    "tags": [
      "rag-pipeline",
      "orchestration",
      "opentelemetry",
      "enterprise-search"
    ],
    "elo": 1290,
    "eloCi": "±18",
    "winRate": 85,
    "starsDisplay": "26.7k",
    "weeklyStars": 11411,
    "weeklyGrowth": "+55.1%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-84%",
    "tokensNum": 84,
    "logoUrl": "https://github.com/deepset-ai.png",
    "githubUrl": "https://github.com/deepset-ai/haystack",
    "rank": 139
  },
  {
    "id": "peft",
    "name": "Hugging Face PEFT",
    "repo": "huggingface/peft",
    "author": "huggingface",
    "persona": "STEM & AI Researchers",
    "targetRole": "Model Fine-Tuning Engineers",
    "stars": 21762,
    "wGrowth": 31.9,
    "mGrowth": 46,
    "website": "https://huggingface.co/docs/peft",
    "traffic": "940k/mo",
    "searchVol": "160,000/mo",
    "latency": 14,
    "tokens": 97,
    "desc": "Parameter-Efficient Fine-Tuning library for efficiently adapting pre-trained models to various downstream applications.",
    "useCase": "Fine-tune 70B parameter LLMs on single consumer GPUs using LoRA, QLoRA, and Prefix Tuning.",
    "arch": "Python, PyTorch, LoRA, AdaLoRA, Prompt Tuning.",
    "tags": [
      "lora",
      "qlora",
      "fine-tuning",
      "parameter-efficient"
    ],
    "starsDisplay": "21.8k",
    "weeklyStars": 7110,
    "weeklyGrowth": "+31.9%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "trafficNum": 940000,
    "searchVolNum": 160000,
    "searchTrend": "+112%",
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/peft",
    "rank": 140
  },
  {
    "id": "surrealdb",
    "name": "SurrealDB",
    "repo": "surrealdb/surrealdb",
    "author": "surrealdb",
    "persona": "Founders & Solo Creators",
    "targetRole": "Multi-Model Cloud Architects",
    "stars": 33108,
    "wGrowth": 13.4,
    "mGrowth": 46,
    "website": "https://surrealdb.com",
    "traffic": "780k/mo",
    "searchVol": "130,000/mo",
    "latency": 18,
    "tokens": 93,
    "desc": "A scalable, distributed, collaborative, document-graph database for web, mobile, serverless, and cloud.",
    "useCase": "Store graph relationships, relational records, and vector embeddings in a single unified multi-model database.",
    "arch": "Rust, TiKV storage engine, SurrealQL, WebSocket sync.",
    "tags": [
      "multi-model",
      "graph-database",
      "rust",
      "realtime"
    ],
    "starsDisplay": "33.1k",
    "weeklyStars": 7178,
    "weeklyGrowth": "+13.4%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "trafficNum": 780000,
    "searchVolNum": 130000,
    "searchTrend": "+112%",
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "elo": 1301,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/surrealdb.png",
    "githubUrl": "https://github.com/surrealdb/surrealdb",
    "rank": 141
  },
  {
    "id": "deepspeed-moe",
    "name": "DeepSpeed-MoE",
    "repo": "microsoft/DeepSpeed-MII",
    "author": "microsoft",
    "persona": "STEM & AI Researchers",
    "targetRole": "MoE Scaling Engineers",
    "stars": 2114,
    "wGrowth": 11.2,
    "mGrowth": 46,
    "website": "https://github.com/microsoft/DeepSpeed-MII",
    "traffic": "280k/mo",
    "searchVol": "42,000/mo",
    "latency": 15,
    "tokens": 96,
    "desc": "Fast and low-latency inference system for Mixture of Experts (MoE) foundation models.",
    "useCase": "Deploy sparse MoE models like DeepSeek-V3 and Mixtral with high throughput and low memory footprint.",
    "arch": "C++, CUDA, ZeRO-Inference, PyTorch.",
    "tags": [
      "moe",
      "deepspeed",
      "fast-inference",
      "microsoft"
    ],
    "starsDisplay": "2.1k",
    "weeklyStars": 504,
    "weeklyGrowth": "+11.2%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "trafficNum": 280000,
    "searchVolNum": 42000,
    "searchTrend": "+112%",
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1298,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/DeepSpeed-MII",
    "rank": 142
  },
  {
    "id": "vespa",
    "name": "Vespa",
    "repo": "vespa-engine/vespa",
    "author": "vespa-engine",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Billion-Scale Search Architects",
    "stars": 7118,
    "wGrowth": 20.6,
    "mGrowth": 46,
    "website": "https://vespa.ai",
    "traffic": "520k/mo",
    "searchVol": "88,000/mo",
    "latency": 14,
    "tokens": 97,
    "desc": "The open big data serving engine for vector search, lexical search, and machine-learned ranking on billions of docs.",
    "useCase": "Serve combined neural vector ranking and BM25 lexical search at tens of thousands of queries per second.",
    "arch": "C++, Java container, Real-time indexing, Tensor computation.",
    "tags": [
      "billion-scale-search",
      "hybrid-search",
      "tensor-engine",
      "vespa"
    ],
    "starsDisplay": "7.1k",
    "weeklyStars": 1879,
    "weeklyGrowth": "+20.6%",
    "monthlyGrowth": "+46%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 46,
    "trafficNum": 520000,
    "searchVolNum": 88000,
    "searchTrend": "+112%",
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "elo": 1298,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/vespa-engine.png",
    "githubUrl": "https://github.com/vespa-engine/vespa",
    "rank": 143
  },
  {
    "id": "autogptq",
    "name": "AutoGPTQ",
    "repo": "AutoGPTQ/AutoGPTQ",
    "author": "AutoGPTQ",
    "persona": "STEM & AI Researchers",
    "targetRole": "GPTQ Quantization Engineers",
    "stars": 5065,
    "wGrowth": 11.2,
    "mGrowth": 47,
    "website": "https://github.com/AutoGPTQ/AutoGPTQ",
    "traffic": "440k/mo",
    "searchVol": "76,000/mo",
    "latency": 11,
    "tokens": 98,
    "desc": "An easy-to-use LLMs quantization package with user-friendly APIs, based on GPTQ algorithm.",
    "useCase": "Compress Hugging Face foundation models to 4-bit and 3-bit weights for low-VRAM deployment.",
    "arch": "Python, PyTorch, CUDA, Triton kernels, Marlin backend.",
    "tags": [
      "gptq",
      "post-training-quantization",
      "low-vram",
      "compression"
    ],
    "starsDisplay": "5.1k",
    "weeklyStars": 571,
    "weeklyGrowth": "+11.2%",
    "monthlyGrowth": "+47%",
    "wGrowthNum": 11.2,
    "mGrowthNum": 47,
    "trafficNum": 440000,
    "searchVolNum": 76000,
    "searchTrend": "+112%",
    "latencyMs": "11ms",
    "latencyNum": 11,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "elo": 1298,
    "eloCi": "±18",
    "winRate": 86,
    "logoUrl": "https://github.com/AutoGPTQ.png",
    "githubUrl": "https://github.com/AutoGPTQ/AutoGPTQ",
    "rank": 144
  },
  {
    "id": "turbopack",
    "name": "Turbopack",
    "repo": "vercel/turbopack",
    "author": "vercel",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Incremental Bundler Specialists",
    "stars": 26542,
    "starsDisplay": "26.5k",
    "weeklyStars": 2946,
    "weeklyGrowth": "+11.1%",
    "monthlyGrowth": "+42%",
    "wGrowthNum": 11.1,
    "mGrowthNum": 42,
    "traffic": "678k/mo",
    "trafficNum": 677789,
    "searchVol": "73,000/mo",
    "searchVolNum": 73196,
    "searchTrend": "+111%",
    "latencyMs": "9ms",
    "latencyNum": 16,
    "tokensSaved": "-90%",
    "tokensNum": 95,
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "website": "https://github.com/vercel/turbopack",
    "githubUrl": "https://github.com/vercel/turbopack",
    "logoUrl": "https://github.com/vercel.png",
    "desc": "High-impact open-source tool empowering incremental bundler specialists with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical incremental-build infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "incremental-build",
      "rust",
      "nextjs-bundler",
      "vercel"
    ],
    "rank": 145
  },
  {
    "id": "lucide",
    "name": "Lucide Icons",
    "repo": "lucide-icons/lucide",
    "author": "lucide-icons",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Design Engineers & Founders",
    "stars": 24885,
    "wGrowth": 31.7,
    "mGrowth": 44,
    "website": "https://lucide.dev",
    "traffic": "2.4M/mo",
    "trafficNum": 2400000,
    "searchVol": "290,000/mo",
    "searchVolNum": 290000,
    "searchTrend": "+160%",
    "latency": 8,
    "tokens": 96,
    "desc": "Beautiful & consistent open-source icon set. A community-maintained fork of Feather Icons with 1,400+ vectors.",
    "useCase": "Standardize iconography across React, Vue, Svelte, and native SVG with consistent 24x24 grid lines.",
    "arch": "SVG vector format, Treeshakeable ES modules, Zero external dependencies.",
    "tags": [
      "icons",
      "svg",
      "design-system",
      "feather-fork"
    ],
    "elo": 1320,
    "eloCi": "±18",
    "winRate": 88,
    "logoUrl": "https://github.com/lucide-icons.png",
    "githubUrl": "https://github.com/lucide-icons/lucide",
    "starsDisplay": "24.9k",
    "weeklyStars": 8064,
    "weeklyGrowth": "+31.7%",
    "monthlyGrowth": "+44.0%",
    "wGrowthNum": 11,
    "mGrowthNum": 44,
    "latencyMs": "8ms",
    "latencyNum": 8,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "rank": 146
  },
  {
    "id": "devika",
    "name": "Devika Agent",
    "repo": "stitionai/devika",
    "author": "stitionai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Autonomous Developers",
    "stars": 19554,
    "wGrowth": 11,
    "mGrowth": 41.8,
    "website": "https://github.com/stitionai/devika",
    "traffic": "990,000/mo",
    "trafficNum": 990000,
    "searchVol": "198,000/mo",
    "searchVolNum": 198000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Devika Agent powering production AI workloads.",
    "useCase": "Integrate Devika Agent into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1324,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "19.6k",
    "weeklyStars": 2178,
    "weeklyGrowth": "+11.0%",
    "monthlyGrowth": "+41.8%",
    "wGrowthNum": 11,
    "mGrowthNum": 41.8,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/stitionai.png",
    "githubUrl": "https://github.com/stitionai/devika",
    "rank": 147
  },
  {
    "id": "semantic-kernel",
    "name": "Semantic Kernel",
    "repo": "microsoft/semantic-kernel",
    "author": "microsoft",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Enterprise SDK Architects",
    "stars": 28628,
    "wGrowth": 20.3,
    "mGrowth": 44,
    "website": "https://learn.microsoft.com/semantic-kernel",
    "traffic": "890k/mo",
    "trafficNum": 890000,
    "searchVol": "150,000/mo",
    "searchVolNum": 150000,
    "searchTrend": "+115%",
    "latency": 26,
    "tokens": 86,
    "desc": "Microsoft SDK that easily integrates LLMs like OpenAI, Azure, and Hugging Face into conventional C#, Python, and Java.",
    "useCase": "Embed enterprise agent plugins and memory abstractions into existing C# .NET or Java enterprise services.",
    "arch": "C#, Python, Java, OpenTelemetry, Native memory connectors.",
    "tags": [
      "microsoft",
      "dotnet-ai",
      "enterprise-agent",
      "plugin-architecture"
    ],
    "elo": 1295,
    "eloCi": "±18",
    "winRate": 85,
    "starsDisplay": "28.6k",
    "weeklyStars": 7422,
    "weeklyGrowth": "+20.3%",
    "monthlyGrowth": "+44%",
    "wGrowthNum": 10.9,
    "mGrowthNum": 44,
    "latencyMs": "26ms",
    "latencyNum": 26,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/semantic-kernel",
    "rank": 148
  },
  {
    "id": "browser-gym",
    "name": "BrowserGym",
    "repo": "ServiceNow/BrowserGym",
    "author": "ServiceNow",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Web Agent Benchmark Specialists",
    "stars": 1389,
    "starsDisplay": "1.4k",
    "weeklyStars": 1319,
    "weeklyGrowth": "+10.9%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.9,
    "mGrowthNum": 41,
    "traffic": "452k/mo",
    "trafficNum": 452103,
    "searchVol": "74,000/mo",
    "searchVolNum": 74263,
    "searchTrend": "+109%",
    "latencyMs": "18ms",
    "latencyNum": 10,
    "tokensSaved": "-96%",
    "tokensNum": 95,
    "elo": 1297,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/ServiceNow/BrowserGym",
    "githubUrl": "https://github.com/ServiceNow/BrowserGym",
    "logoUrl": "https://github.com/ServiceNow.png",
    "desc": "High-impact open-source repository empowering web agent benchmark specialists with production capabilities.",
    "useCase": "Used by modern agentic teams to scale browser-gym infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "browser-gym",
      "web-agent-eval",
      "servicenow",
      "benchmark"
    ],
    "rank": 149
  },
  {
    "id": "smolagents",
    "name": "Smolagents",
    "repo": "huggingface/smolagents",
    "author": "huggingface",
    "persona": "AI Ops & Tooling",
    "targetRole": "Agent Researchers",
    "stars": 29714,
    "wGrowth": 109.3,
    "mGrowth": 46.5,
    "website": "https://huggingface.co",
    "traffic": "2.4M/mo",
    "trafficNum": 2400000,
    "searchVol": "48,000/mo",
    "searchVolNum": 48000,
    "searchTrend": "+108%",
    "latency": 50,
    "tokens": 91,
    "desc": "A ridiculously simple library for building agents in ~1,000 lines of Python.",
    "useCase": "Deploy lightweight autonomous agents writing code directly to solve complex tasks.",
    "arch": "Python, Transformers, Sandboxed execution.",
    "tags": [
      "agents",
      "huggingface",
      "minimalist"
    ],
    "rank": 150,
    "elo": 1214,
    "eloCi": "±19",
    "winRate": 75,
    "starsDisplay": "29.7k",
    "weeklyStars": 17047,
    "weeklyGrowth": "+109.3%",
    "monthlyGrowth": "+46.5%",
    "wGrowthNum": 10.8,
    "mGrowthNum": 46.5,
    "latencyMs": "50ms",
    "latencyNum": 50,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/smolagents"
  },
  {
    "id": "typebot",
    "name": "Typebot",
    "repo": "baptisteArno/typebot.io",
    "author": "baptisteArno",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Growth Hackers & Marketers",
    "stars": 10504,
    "wGrowth": 10.8,
    "mGrowth": 42,
    "website": "https://typebot.io",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "180,000/mo",
    "searchVolNum": 180000,
    "searchTrend": "+145%",
    "latency": 22,
    "tokens": 90,
    "desc": "Open-source conversational form builder. Embed conversational agent flows on web apps to collect leads and answers.",
    "useCase": "Convert static lead capture landing pages into high-converting, personalized conversational qualification loops.",
    "arch": "TypeScript, Next.js, Prisma, PostgreSQL, Tailwind CSS.",
    "tags": [
      "lead-generation",
      "conversational-forms",
      "growth",
      "nextjs"
    ],
    "elo": 1285,
    "eloCi": "±21",
    "winRate": 83,
    "logoUrl": "https://github.com/baptisteArno.png",
    "githubUrl": "https://github.com/baptisteArno/typebot.io",
    "starsDisplay": "10.5k",
    "weeklyStars": 2138,
    "weeklyGrowth": "+10.8%",
    "monthlyGrowth": "+42.0%",
    "wGrowthNum": 10.8,
    "mGrowthNum": 42,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "rank": 151
  },
  {
    "id": "tiktoken",
    "name": "OpenAI Tiktoken",
    "repo": "openai/tiktoken",
    "author": "openai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Prompt Engineers",
    "stars": 19399,
    "wGrowth": 43.7,
    "mGrowth": 41,
    "website": "https://github.com/openai/tiktoken",
    "traffic": "675,000/mo",
    "trafficNum": 675000,
    "searchVol": "135,000/mo",
    "searchVolNum": 135000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool OpenAI Tiktoken powering production AI workloads.",
    "useCase": "Integrate OpenAI Tiktoken into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1323,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "19.4k",
    "weeklyStars": 7357,
    "weeklyGrowth": "+43.7%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.8,
    "mGrowthNum": 41,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/openai.png",
    "githubUrl": "https://github.com/openai/tiktoken",
    "rank": 152
  },
  {
    "id": "flyte",
    "name": "Flyte",
    "repo": "flyteorg/flyte",
    "author": "flyteorg",
    "persona": "STEM & AI Researchers",
    "targetRole": "Orchestration Platform Devs",
    "stars": 7643,
    "starsDisplay": "7.6k",
    "weeklyStars": 1573,
    "weeklyGrowth": "+10.8%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.8,
    "mGrowthNum": 41,
    "traffic": "643k/mo",
    "trafficNum": 642807,
    "searchVol": "149,000/mo",
    "searchVolNum": 148597,
    "searchTrend": "+108%",
    "latencyMs": "26ms",
    "latencyNum": 13,
    "tokensSaved": "-98%",
    "tokensNum": 89,
    "elo": 1297,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/flyteorg/flyte",
    "githubUrl": "https://github.com/flyteorg/flyte",
    "logoUrl": "https://github.com/flyteorg.png",
    "desc": "High-impact open-source repository empowering orchestration platform devs with production capabilities.",
    "useCase": "Used by modern agentic teams to scale large-scale-orchestration infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "large-scale-orchestration",
      "kubernetes-ml",
      "workflow",
      "flyte"
    ],
    "rank": 153
  },
  {
    "id": "tauri",
    "name": "Tauri",
    "repo": "tauri-apps/tauri",
    "author": "tauri-apps",
    "persona": "Founders & Solo Creators",
    "targetRole": "Lightweight Desktop Architects",
    "stars": 111631,
    "starsDisplay": "111.6k",
    "weeklyStars": 92983,
    "weeklyGrowth": "+434.6%",
    "monthlyGrowth": "+40%",
    "wGrowthNum": 10.7,
    "mGrowthNum": 40,
    "traffic": "624k/mo",
    "trafficNum": 623843,
    "searchVol": "153,000/mo",
    "searchVolNum": 152890,
    "searchTrend": "+107%",
    "latencyMs": "10ms",
    "latencyNum": 20,
    "tokensSaved": "-97%",
    "tokensNum": 90,
    "elo": 1296,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/tauri-apps/tauri",
    "githubUrl": "https://github.com/tauri-apps/tauri",
    "logoUrl": "https://github.com/tauri-apps.png",
    "desc": "High-impact open-source tool empowering lightweight desktop architects with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical desktop-apps infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "desktop-apps",
      "rust-backend",
      "electron-alternative",
      "tauri"
    ],
    "rank": 154,
    "wGrowth": 434.6
  },
  {
    "id": "redpanda",
    "name": "Redpanda",
    "repo": "redpanda-data/redpanda",
    "author": "redpanda-data",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Streaming Data Architects",
    "stars": 12602,
    "starsDisplay": "12.6k",
    "weeklyStars": 2168,
    "weeklyGrowth": "+10.6%",
    "monthlyGrowth": "+40%",
    "wGrowthNum": 10.6,
    "mGrowthNum": 40,
    "traffic": "1134k/mo",
    "trafficNum": 1133669,
    "searchVol": "156,000/mo",
    "searchVolNum": 156013,
    "searchTrend": "+106%",
    "latencyMs": "10ms",
    "latencyNum": 23,
    "tokensSaved": "-96%",
    "tokensNum": 89,
    "elo": 1296,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/redpanda-data/redpanda",
    "githubUrl": "https://github.com/redpanda-data/redpanda",
    "logoUrl": "https://github.com/redpanda-data.png",
    "desc": "High-impact open-source repository empowering streaming data architects with production capabilities.",
    "useCase": "Used by modern agentic teams to scale kafka-alternative infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "kafka-alternative",
      "c-plus-plus",
      "zero-jvm",
      "streaming"
    ],
    "rank": 155
  },
  {
    "id": "shell-genie",
    "name": "Shell Genie",
    "repo": "euclio/shell-genie",
    "author": "euclio",
    "persona": "Founders & Solo Creators",
    "targetRole": "Terminal Command Translators",
    "stars": 35683,
    "starsDisplay": "35.7k",
    "weeklyStars": 3782,
    "weeklyGrowth": "+10.6%",
    "monthlyGrowth": "+40%",
    "wGrowthNum": 10.6,
    "mGrowthNum": 40,
    "traffic": "470k/mo",
    "trafficNum": 469821,
    "searchVol": "106,000/mo",
    "searchVolNum": 105840,
    "searchTrend": "+106%",
    "latencyMs": "19ms",
    "latencyNum": 15,
    "tokensSaved": "-91%",
    "tokensNum": 94,
    "elo": 1297,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/euclio/shell-genie",
    "githubUrl": "https://github.com/euclio/shell-genie",
    "logoUrl": "https://github.com/euclio.png",
    "desc": "High-impact open-source repository empowering terminal command translators with production capabilities.",
    "useCase": "Used by modern agentic teams to scale terminal-genie infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "terminal-genie",
      "cli-helper",
      "natural-language-bash",
      "python"
    ],
    "rank": 156
  },
  {
    "id": "pr-agent",
    "name": "Codium PR-Agent",
    "repo": "Codium-ai/pr-agent",
    "author": "Codium-ai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "DevOps & Engineering Leads",
    "stars": 13291,
    "wGrowth": 10.5,
    "mGrowth": 42,
    "website": "https://qodo.ai",
    "traffic": "740k/mo",
    "trafficNum": 740000,
    "searchVol": "110,000/mo",
    "searchVolNum": 110000,
    "searchTrend": "+125%",
    "latency": 26,
    "tokens": 86,
    "desc": "Autonomous AI-powered tool for automated pull request analysis, feedback, suggestions, and auto-generated descriptions.",
    "useCase": "Automatically audit pull requests for security vulnerabilities, race conditions, and test coverage before human merge.",
    "arch": "Python, GitHub Actions, GitLab CI, Bitbucket integration.",
    "tags": [
      "code-review",
      "pull-request",
      "ci-cd",
      "automation"
    ],
    "elo": 1300,
    "eloCi": "±20",
    "winRate": 85,
    "logoUrl": "https://github.com/Codium-ai.png",
    "githubUrl": "https://github.com/Codium-ai/pr-agent",
    "starsDisplay": "13.3k",
    "weeklyStars": 1901,
    "weeklyGrowth": "+10.5%",
    "monthlyGrowth": "+42.0%",
    "wGrowthNum": 10.5,
    "mGrowthNum": 42,
    "latencyMs": "26ms",
    "latencyNum": 26,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "rank": 157
  },
  {
    "id": "instantid",
    "name": "InstantID",
    "repo": "InstantID/InstantID",
    "author": "InstantID",
    "persona": "Music & Audio Creators",
    "targetRole": "Creative Designers",
    "stars": 11994,
    "wGrowth": 10.5,
    "mGrowth": 39.9,
    "website": "https://github.com/InstantID/InstantID",
    "traffic": "910,000/mo",
    "trafficNum": 910000,
    "searchVol": "182,000/mo",
    "searchVolNum": 182000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool InstantID powering production AI workloads.",
    "useCase": "Integrate InstantID into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1322,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "12.0k",
    "weeklyStars": 1911,
    "weeklyGrowth": "+10.5%",
    "monthlyGrowth": "+39.9%",
    "wGrowthNum": 10.5,
    "mGrowthNum": 39.9,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/InstantID.png",
    "githubUrl": "https://github.com/InstantID/InstantID",
    "rank": 158
  },
  {
    "id": "audioldm",
    "name": "AudioLDM",
    "repo": "haoheliu/AudioLDM",
    "author": "haoheliu",
    "persona": "Music & Audio Creators",
    "targetRole": "Sound Effects Designers",
    "stars": 2911,
    "wGrowth": 10.5,
    "mGrowth": 41,
    "website": "https://audioldm.github.io",
    "traffic": "290k/mo",
    "searchVol": "48,000/mo",
    "latency": 45,
    "tokens": 88,
    "desc": "Text-to-audio generation using latent diffusion models for sound effects, music, and Foley art.",
    "useCase": "Generate bespoke cinematic sound effects and background Foley audio from text descriptions.",
    "arch": "PyTorch, Latent Diffusion, CLAP audio embeddings.",
    "tags": [
      "sound-effects",
      "latent-diffusion",
      "foley-ai",
      "cinematic-audio"
    ],
    "starsDisplay": "2.9k",
    "weeklyStars": 473,
    "weeklyGrowth": "+10.5%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.5,
    "mGrowthNum": 41,
    "trafficNum": 290000,
    "searchVolNum": 48000,
    "searchTrend": "+105%",
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "elo": 1293,
    "eloCi": "±18",
    "winRate": 85,
    "logoUrl": "https://github.com/haoheliu.png",
    "githubUrl": "https://github.com/haoheliu/AudioLDM",
    "rank": 159
  },
  {
    "id": "tgi",
    "name": "Text Generation Inference (TGI)",
    "repo": "huggingface/text-generation-inference",
    "author": "huggingface",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Hugging Face Serving Architects",
    "stars": 10880,
    "wGrowth": 10.5,
    "mGrowth": 43,
    "website": "https://github.com/huggingface/text-generation-inference",
    "traffic": "950k/mo",
    "searchVol": "160,000/mo",
    "latency": 14,
    "tokens": 96,
    "desc": "Production-grade, highly optimized engine for deploying and serving Large Language Models at scale.",
    "useCase": "Deploy Hugging Face open weights with FlashAttention, continuous batching, and tensor parallelism on Kubernetes.",
    "arch": "Rust router, Python gRPC engine, CUDA kernels, FlashAttention.",
    "tags": [
      "tgi",
      "huggingface-serving",
      "continuous-batching",
      "production-llm"
    ],
    "starsDisplay": "10.9k",
    "weeklyStars": 1751,
    "weeklyGrowth": "+10.5%",
    "monthlyGrowth": "+43%",
    "wGrowthNum": 10.5,
    "mGrowthNum": 43,
    "trafficNum": 950000,
    "searchVolNum": 160000,
    "searchTrend": "+105%",
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1294,
    "eloCi": "±18",
    "winRate": 85,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/text-generation-inference",
    "rank": 160
  },
  {
    "id": "fabric",
    "name": "Fabric",
    "repo": "danielmiessler/fabric",
    "author": "danielmiessler",
    "persona": "Founders & Solo Creators",
    "targetRole": "Human-AI Augmentation Leads",
    "stars": 44170,
    "starsDisplay": "44.2k",
    "weeklyStars": 26944,
    "weeklyGrowth": "+129.5%",
    "monthlyGrowth": "+39%",
    "wGrowthNum": 10.5,
    "mGrowthNum": 39,
    "traffic": "783k/mo",
    "trafficNum": 782813,
    "searchVol": "124,000/mo",
    "searchVolNum": 124088,
    "searchTrend": "+105%",
    "latencyMs": "10ms",
    "latencyNum": 19,
    "tokensSaved": "-93%",
    "tokensNum": 89,
    "elo": 1295,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/danielmiessler/fabric",
    "githubUrl": "https://github.com/danielmiessler/fabric",
    "logoUrl": "https://github.com/danielmiessler.png",
    "desc": "High-impact open-source repository empowering human-ai augmentation leads with production capabilities.",
    "useCase": "Used by modern agentic teams to scale prompt-patterns infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "prompt-patterns",
      "human-augmentation",
      "cli-workflow",
      "fabric"
    ],
    "rank": 161,
    "wGrowth": 129.5
  },
  {
    "id": "radix-ui",
    "name": "Radix UI Primitives",
    "repo": "radix-ui/primitives",
    "author": "radix-ui",
    "persona": "Founders & Indie Hackers",
    "targetRole": "UI Engineers",
    "stars": 19366,
    "wGrowth": 15.3,
    "mGrowth": 39.5,
    "website": "https://github.com/radix-ui/primitives",
    "traffic": "840,000/mo",
    "trafficNum": 840000,
    "searchVol": "168,000/mo",
    "searchVolNum": 168000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Radix UI Primitives powering production AI workloads.",
    "useCase": "Integrate Radix UI Primitives into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1322,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "19.4k",
    "weeklyStars": 4313,
    "weeklyGrowth": "+15.3%",
    "monthlyGrowth": "+39.5%",
    "wGrowthNum": 10.4,
    "mGrowthNum": 39.5,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/radix-ui.png",
    "githubUrl": "https://github.com/radix-ui/primitives",
    "rank": 162
  },
  {
    "id": "flash-attention",
    "name": "FlashAttention",
    "repo": "Dao-AILab/flash-attention",
    "author": "Dao-AILab",
    "persona": "STEM & AI Researchers",
    "targetRole": "Kernel Optimization Specialists",
    "stars": 25098,
    "wGrowth": 41,
    "mGrowth": 41,
    "website": "https://github.com/Dao-AILab/flash-attention",
    "traffic": "680k/mo",
    "searchVol": "110,000/mo",
    "latency": 8,
    "tokens": 99,
    "desc": "Fast and memory-efficient exact attention with IO-awareness, unlocking massive LLM context windows.",
    "useCase": "Accelerate Transformer attention 2-4x while reducing memory footprint from quadratic to linear.",
    "arch": "CUDA, C++, GPU SRAM tiling, IO optimization.",
    "tags": [
      "fast-attention",
      "cuda-speed",
      "context-window",
      "transformer-kernel"
    ],
    "starsDisplay": "25.1k",
    "weeklyStars": 9149,
    "weeklyGrowth": "+41%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.4,
    "mGrowthNum": 41,
    "trafficNum": 680000,
    "searchVolNum": 110000,
    "searchTrend": "+104%",
    "latencyMs": "8ms",
    "latencyNum": 8,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "elo": 1294,
    "eloCi": "±18",
    "winRate": 85,
    "logoUrl": "https://github.com/Dao-AILab.png",
    "githubUrl": "https://github.com/Dao-AILab/flash-attention",
    "rank": 163
  },
  {
    "id": "esbuild",
    "name": "esbuild",
    "repo": "evanw/esbuild",
    "author": "evanw",
    "persona": "Autonomous Software Engineers",
    "targetRole": "High-Speed Bundler Architects",
    "stars": 40073,
    "starsDisplay": "40.1k",
    "weeklyStars": 15165,
    "weeklyGrowth": "+44.3%",
    "monthlyGrowth": "+39%",
    "wGrowthNum": 10.3,
    "mGrowthNum": 39,
    "traffic": "594k/mo",
    "trafficNum": 594208,
    "searchVol": "149,000/mo",
    "searchVolNum": 148926,
    "searchTrend": "+103%",
    "latencyMs": "9ms",
    "latencyNum": 12,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "elo": 1294,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/evanw/esbuild",
    "githubUrl": "https://github.com/evanw/esbuild",
    "logoUrl": "https://github.com/evanw.png",
    "desc": "High-impact open-source tool empowering high-speed bundler architects with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical bundler infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "bundler",
      "golang",
      "fast-build",
      "esbuild"
    ],
    "rank": 164,
    "wGrowth": 44.3
  },
  {
    "id": "postiz",
    "name": "Postiz",
    "repo": "gitroomhq/postiz-app",
    "author": "gitroomhq",
    "persona": "Marketers & Growth",
    "targetRole": "Social Media Managers",
    "stars": 36809,
    "wGrowth": 133,
    "mGrowth": 44,
    "website": "https://postiz.com",
    "traffic": "420k/mo",
    "trafficNum": 420000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+110%",
    "latency": 45,
    "tokens": 88,
    "desc": "Open-source social media scheduling tool with AI generation, multi-account calendar management, and thread analytics.",
    "useCase": "Schedule 30 days of Twitter/X threads, LinkedIn carousels, and Reddit updates in one sitting with automated best-time posting.",
    "arch": "Next.js, Prisma, PostgreSQL, BullMQ, Tailwind CSS.",
    "tags": [
      "social-scheduling",
      "x-threads",
      "linkedin",
      "viral-growth"
    ],
    "rank": 165,
    "elo": 1212,
    "eloCi": "±20",
    "winRate": 75,
    "starsDisplay": "36.8k",
    "weeklyStars": 22620,
    "weeklyGrowth": "+133%",
    "monthlyGrowth": "+44.0%",
    "wGrowthNum": 10.2,
    "mGrowthNum": 44,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/gitroomhq.png",
    "githubUrl": "https://github.com/gitroomhq/postiz-app"
  },
  {
    "id": "anthropic-quickstarts",
    "name": "Anthropic Computer Use",
    "repo": "anthropics/anthropic-quickstarts",
    "author": "anthropics",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Claude AI Engineers",
    "stars": 17811,
    "wGrowth": "+10.2%",
    "mGrowth": "+41.3%",
    "website": "https://docs.anthropic.com",
    "traffic": "2.1M/mo",
    "trafficNum": 2100000,
    "searchVol": "120,000/mo",
    "searchVolNum": 120000,
    "searchTrend": "+92% (Surging)",
    "latency": "95ms",
    "tokens": "-55%",
    "desc": "Official reference implementations for Claude 3.7 / 3.5 Sonnet computer use, Model Context Protocol (MCP) clients, and automated financial agents.",
    "useCase": "Equipping Claude agents with desktop OS GUI control, keyboard/mouse actions, and standardized MCP tool connectors.",
    "arch": "Docker, Python, Claude Computer Use API, MCP client bridge.",
    "tags": [
      "Computer Use",
      "Claude 3.7",
      "MCP",
      "GUI Automation",
      "Anthropic"
    ],
    "wGrowthNum": 10.2,
    "mGrowthNum": 41.3,
    "latencyNum": 95,
    "tokensNum": 55,
    "weeklyStars": 2692,
    "weeklyGrowth": "+10.2%",
    "monthlyGrowth": "+41.3%",
    "latencyMs": "95ms",
    "tokensSaved": "-55%",
    "starsDisplay": "17.8k",
    "logoUrl": "https://github.com/anthropics.png",
    "githubUrl": "https://github.com/anthropics/anthropic-quickstarts",
    "rank": 166,
    "elo": 1233,
    "eloCi": "±21",
    "winRate": 78
  },
  {
    "id": "flowise",
    "name": "Flowise",
    "repo": "FlowiseAI/Flowise",
    "author": "FlowiseAI",
    "persona": "Founders & Indie Hackers",
    "targetRole": "No-Code & Rapid Prototypers",
    "stars": 55487,
    "wGrowth": 60.8,
    "mGrowth": 40,
    "website": "https://flowiseai.com",
    "traffic": "1.6M/mo",
    "trafficNum": 1600000,
    "searchVol": "230,000/mo",
    "searchVolNum": 230000,
    "searchTrend": "+130%",
    "latency": 28,
    "tokens": 88,
    "desc": "Drag & drop UI to build customized LLM flows using LangChain and LlamaIndex. Self-hosted and developer-friendly.",
    "useCase": "Prototype and deploy production LangChain flows as REST API endpoints in minutes without writing backend glue.",
    "arch": "TypeScript, React, Node.js, LangChain.js, SQLite / MySQL.",
    "tags": [
      "drag-and-drop",
      "langchain",
      "visual-ai",
      "rapid-prototyping"
    ],
    "elo": 1295,
    "eloCi": "±20",
    "winRate": 84,
    "logoUrl": "https://github.com/FlowiseAI.png",
    "githubUrl": "https://github.com/FlowiseAI/Flowise",
    "starsDisplay": "55.5k",
    "weeklyStars": 24506,
    "weeklyGrowth": "+60.8%",
    "monthlyGrowth": "+40.0%",
    "wGrowthNum": 10.2,
    "mGrowthNum": 40,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 167
  },
  {
    "id": "llava",
    "name": "LLaVA",
    "repo": "haotian-liu/LLaVA",
    "author": "haotian-liu",
    "persona": "Researchers & STEM",
    "targetRole": "Multimodal Researchers",
    "stars": 25059,
    "wGrowth": 10.2,
    "mGrowth": 38.8,
    "website": "https://github.com/haotian-liu/LLaVA",
    "traffic": "1,140,000/mo",
    "trafficNum": 1140000,
    "searchVol": "228,000/mo",
    "searchVolNum": 228000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool LLaVA powering production AI workloads.",
    "useCase": "Integrate LLaVA into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1321,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "25.1k",
    "weeklyStars": 4585,
    "weeklyGrowth": "+10.2%",
    "monthlyGrowth": "+38.8%",
    "wGrowthNum": 10.2,
    "mGrowthNum": 38.8,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/haotian-liu.png",
    "githubUrl": "https://github.com/haotian-liu/LLaVA",
    "rank": 168
  },
  {
    "id": "gorilla",
    "name": "Gorilla",
    "repo": "ShishirPatil/gorilla",
    "author": "ShishirPatil",
    "persona": "STEM & AI Researchers",
    "targetRole": "API Calling Researchers",
    "stars": 13042,
    "starsDisplay": "13.0k",
    "weeklyStars": 1442,
    "weeklyGrowth": "+10.2%",
    "monthlyGrowth": "+38%",
    "wGrowthNum": 10.2,
    "mGrowthNum": 38,
    "traffic": "758k/mo",
    "trafficNum": 758384,
    "searchVol": "154,000/mo",
    "searchVolNum": 154496,
    "searchTrend": "+102%",
    "latencyMs": "13ms",
    "latencyNum": 14,
    "tokensSaved": "-93%",
    "tokensNum": 97,
    "elo": 1292,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/ShishirPatil/gorilla",
    "githubUrl": "https://github.com/ShishirPatil/gorilla",
    "logoUrl": "https://github.com/ShishirPatil.png",
    "desc": "High-impact open-source repository empowering api calling researchers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale api-calling infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "api-calling",
      "berkeley",
      "function-calling",
      "tool-use"
    ],
    "rank": 169
  },
  {
    "id": "hyperfine",
    "name": "hyperfine",
    "repo": "sharkdp/hyperfine",
    "author": "sharkdp",
    "persona": "Autonomous Software Engineers",
    "targetRole": "CLI Benchmark Engineers",
    "stars": 28953,
    "starsDisplay": "29.0k",
    "weeklyStars": 9653,
    "weeklyGrowth": "+34.7%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 10.2,
    "mGrowthNum": 41,
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "65,000/mo",
    "searchVolNum": 65000,
    "searchTrend": "+102%",
    "latencyMs": "6ms",
    "latencyNum": 6,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "elo": 1310,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/sharkdp/hyperfine",
    "githubUrl": "https://github.com/sharkdp/hyperfine",
    "logoUrl": "https://github.com/sharkdp.png",
    "desc": "A command-line benchmarking tool written in Rust for rigorous performance evaluation.",
    "useCase": "Benchmark agent execution runs, compile times, and script latency with statistical rigor.",
    "arch": "Rust, Statistical outlier detection, Multi-threaded warmups.",
    "tags": [
      "benchmarking",
      "cli-tool",
      "rust-performance",
      "profiling"
    ],
    "rank": 170,
    "wGrowth": 34.7
  },
  {
    "id": "pgvector",
    "name": "pgvector",
    "repo": "pgvector/pgvector",
    "author": "pgvector",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Database & Backend Engineers",
    "stars": 23261,
    "wGrowth": 30.7,
    "mGrowth": 39,
    "website": "https://github.com/pgvector/pgvector",
    "traffic": "920k/mo",
    "trafficNum": 920000,
    "searchVol": "240,000/mo",
    "searchVolNum": 240000,
    "searchTrend": "+155%",
    "latency": 14,
    "tokens": 94,
    "desc": "Open-source vector similarity search extension for PostgreSQL. Native HNSW and IVFFlat indexing.",
    "useCase": "Store embeddings directly alongside relational data in standard PostgreSQL without running dedicated vector SaaS.",
    "arch": "C extension for PostgreSQL, HNSW, IVFFlat, L2 / Cosine / Inner Product distances.",
    "tags": [
      "postgres",
      "vector-search",
      "hnsw",
      "database-extension"
    ],
    "elo": 1335,
    "eloCi": "±18",
    "winRate": 89,
    "logoUrl": "https://github.com/pgvector.png",
    "githubUrl": "https://github.com/pgvector/pgvector",
    "starsDisplay": "23.3k",
    "weeklyStars": 7259,
    "weeklyGrowth": "+30.7%",
    "monthlyGrowth": "+39.0%",
    "wGrowthNum": 10.1,
    "mGrowthNum": 39,
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "rank": 171
  },
  {
    "id": "vector",
    "name": "Vector",
    "repo": "vectordotdev/vector",
    "author": "vectordotdev",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Telemetry Pipeline Engineers",
    "stars": 22672,
    "starsDisplay": "22.7k",
    "weeklyStars": 3199,
    "weeklyGrowth": "+10.0%",
    "monthlyGrowth": "+38%",
    "wGrowthNum": 10,
    "mGrowthNum": 38,
    "traffic": "938k/mo",
    "trafficNum": 938372,
    "searchVol": "46,000/mo",
    "searchVolNum": 46198,
    "searchTrend": "+100%",
    "latencyMs": "26ms",
    "latencyNum": 25,
    "tokensSaved": "-98%",
    "tokensNum": 93,
    "elo": 1293,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/vectordotdev/vector",
    "githubUrl": "https://github.com/vectordotdev/vector",
    "logoUrl": "https://github.com/vectordotdev.png",
    "desc": "High-impact open-source repository empowering telemetry pipeline engineers with production capabilities.",
    "useCase": "Used by modern agentic teams to scale observability infrastructure with verified provenance.",
    "arch": "High-efficiency native pipeline with verifiable benchmark proofs and automated CI.",
    "tags": [
      "observability",
      "log-collector",
      "rust-pipeline",
      "telemetry"
    ],
    "rank": 172
  },
  {
    "id": "nushell",
    "name": "Nushell",
    "repo": "nushell/nushell",
    "author": "nushell",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Structured Data Shell Architects",
    "stars": 40628,
    "starsDisplay": "40.6k",
    "weeklyStars": 16205,
    "weeklyGrowth": "+49.7%",
    "monthlyGrowth": "+38%",
    "wGrowthNum": 10,
    "mGrowthNum": 38,
    "traffic": "1046k/mo",
    "trafficNum": 1045881,
    "searchVol": "97,000/mo",
    "searchVolNum": 96920,
    "searchTrend": "+100%",
    "latencyMs": "15ms",
    "latencyNum": 12,
    "tokensSaved": "-92%",
    "tokensNum": 96,
    "elo": 1292,
    "eloCi": "±18",
    "winRate": 85,
    "website": "https://github.com/nushell/nushell",
    "githubUrl": "https://github.com/nushell/nushell",
    "logoUrl": "https://github.com/nushell.png",
    "desc": "High-impact open-source tool empowering structured data shell architects with ultra-fast performance.",
    "useCase": "Adopted across elite engineering organizations for mission-critical structured-shell infrastructure.",
    "arch": "Engineered for zero-overhead execution with native memory safety and deterministic guarantees.",
    "tags": [
      "structured-shell",
      "table-pipeline",
      "rust-cli",
      "nushell"
    ],
    "rank": 173,
    "wGrowth": 49.7
  },
  {
    "id": "sglang",
    "name": "SGLang",
    "repo": "sgl-project/sglang",
    "author": "sgl-project",
    "persona": "AI Ops & Tooling",
    "targetRole": "Inference Optimization Leads",
    "stars": 36830,
    "wGrowth": 241,
    "mGrowth": 42.4,
    "website": "https://sgl-project.github.io",
    "traffic": "620k/mo",
    "trafficNum": 620000,
    "searchVol": "45,000/mo",
    "searchVolNum": 45000,
    "searchTrend": "+92%",
    "latency": 18,
    "tokens": 99,
    "desc": "High-performance serving framework with RadixAttention and fast structured output.",
    "useCase": "Accelerate complex multi-turn agent execution with multi-call KV-cache reuse.",
    "arch": "RadixAttention, FlashInfer, Python runtime.",
    "tags": [
      "serving",
      "radix-attention",
      "kv-cache"
    ],
    "rank": 174,
    "elo": 1199,
    "eloCi": "±22",
    "winRate": 73,
    "starsDisplay": "36.8k",
    "weeklyStars": 27088,
    "weeklyGrowth": "+241%",
    "monthlyGrowth": "+42.4%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 42.4,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "logoUrl": "https://github.com/sgl-project.png",
    "githubUrl": "https://github.com/sgl-project/sglang"
  },
  {
    "id": "autogen",
    "name": "Microsoft AutoGen",
    "repo": "microsoft/autogen",
    "author": "microsoft",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Research Engineers & AI Architects",
    "stars": 61276,
    "wGrowth": 66.5,
    "mGrowth": 38,
    "website": "https://microsoft.github.io/autogen",
    "traffic": "2.4M/mo",
    "trafficNum": 2400000,
    "searchVol": "310,000/mo",
    "searchVolNum": 310000,
    "searchTrend": "+120%",
    "latency": 42,
    "tokens": 78,
    "desc": "Multi-agent conversation framework enabling next-generation LLM applications with customizable, conversable agents.",
    "useCase": "Simulate multi-agent software engineering councils where agents write, test, debug, and review code collaboratively.",
    "arch": "Python, Event-driven architecture, Docker code execution sandbox.",
    "tags": [
      "multi-agent",
      "microsoft",
      "code-execution",
      "conversational-ai"
    ],
    "elo": 1315,
    "eloCi": "±19",
    "winRate": 86,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/autogen",
    "starsDisplay": "61.3k",
    "weeklyStars": 28082,
    "weeklyGrowth": "+66.5%",
    "monthlyGrowth": "+38.0%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 38,
    "latencyMs": "42ms",
    "latencyNum": 42,
    "tokensSaved": "-78%",
    "tokensNum": 78,
    "rank": 175
  },
  {
    "id": "polars",
    "name": "Polars",
    "repo": "pola-rs/polars",
    "author": "pola-rs",
    "persona": "Researchers & STEM",
    "targetRole": "Data Scientists & Quant Engineers",
    "stars": 39970,
    "wGrowth": 23.4,
    "mGrowth": 38,
    "website": "https://pola.rs",
    "traffic": "2.1M/mo",
    "trafficNum": 2100000,
    "searchVol": "340,000/mo",
    "searchVolNum": 340000,
    "searchTrend": "+145%",
    "latency": 14,
    "tokens": 96,
    "desc": "Blazingly fast DataFrames library implemented in Rust. Multi-threaded query engine, memory-efficient Arrow format.",
    "useCase": "Process multi-gigabyte financial and telemetry datasets 10x-50x faster than Pandas with zero out-of-core errors.",
    "arch": "Rust, Apache Arrow, SIMD vectorization, Lazy query optimizer.",
    "tags": [
      "dataframes",
      "rust",
      "high-performance",
      "pandas-alternative"
    ],
    "elo": 1345,
    "eloCi": "±17",
    "winRate": 90,
    "logoUrl": "https://github.com/pola-rs.png",
    "githubUrl": "https://github.com/pola-rs/polars",
    "starsDisplay": "40.0k",
    "weeklyStars": 10745,
    "weeklyGrowth": "+23.4%",
    "monthlyGrowth": "+38.0%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 38,
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "rank": 176
  },
  {
    "id": "diffusers",
    "name": "HuggingFace Diffusers",
    "repo": "huggingface/diffusers",
    "author": "huggingface",
    "persona": "Music & Audio Creators",
    "targetRole": "Generative Artists",
    "stars": 34679,
    "wGrowth": 26.1,
    "mGrowth": 37.2,
    "website": "https://github.com/huggingface/diffusers",
    "traffic": "1,375,000/mo",
    "trafficNum": 1375000,
    "searchVol": "275,000/mo",
    "searchVolNum": 275000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool HuggingFace Diffusers powering production AI workloads.",
    "useCase": "Integrate HuggingFace Diffusers into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1319,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "34.7k",
    "weeklyStars": 9874,
    "weeklyGrowth": "+26.1%",
    "monthlyGrowth": "+37.2%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 37.2,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/diffusers",
    "rank": 177
  },
  {
    "id": "guidance",
    "name": "Guidance",
    "repo": "guidance-ai/guidance",
    "author": "guidance-ai",
    "persona": "STEM & AI Researchers",
    "targetRole": "Constraint-Driven AI Engineers",
    "stars": 21789,
    "wGrowth": 15.3,
    "mGrowth": 38,
    "website": "https://github.com/guidance-ai/guidance",
    "traffic": "240k/mo",
    "trafficNum": 240000,
    "searchVol": "45,000/mo",
    "searchVolNum": 45000,
    "searchTrend": "+95%",
    "latency": 16,
    "tokens": 95,
    "desc": "A guidance language for controlling large language models with token-level grammars and constrained generation.",
    "useCase": "Guarantee 100% syntactically valid JSON, regex-matching strings, and deterministic choices without retry loops.",
    "arch": "C++ token engine, Python bindings, Context-Free Grammars.",
    "tags": [
      "constrained-decoding",
      "grammar-guarantee",
      "token-level-control",
      "structured-output"
    ],
    "elo": 1285,
    "eloCi": "±19",
    "winRate": 84,
    "starsDisplay": "21.8k",
    "weeklyStars": 4741,
    "weeklyGrowth": "+15.3%",
    "monthlyGrowth": "+38%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 38,
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/guidance-ai.png",
    "githubUrl": "https://github.com/guidance-ai/guidance",
    "rank": 178
  },
  {
    "id": "accelerate",
    "name": "Hugging Face Accelerate",
    "repo": "huggingface/accelerate",
    "author": "huggingface",
    "persona": "STEM & AI Researchers",
    "targetRole": "PyTorch Training Engineers",
    "stars": 9909,
    "wGrowth": 9.8,
    "mGrowth": 39,
    "website": "https://huggingface.co/docs/accelerate",
    "traffic": "890k/mo",
    "searchVol": "140,000/mo",
    "latency": 12,
    "tokens": 98,
    "desc": "Enables the same PyTorch code to be run across any distributed configuration (CPU, multi-GPU, TPU).",
    "useCase": "Train and evaluate models effortlessly across single-GPU workstations or multi-node clusters without rewriting code.",
    "arch": "Python, PyTorch native, FSDP, DeepSpeed integration.",
    "tags": [
      "pytorch-acceleration",
      "multi-gpu",
      "distributed-ml",
      "huggingface"
    ],
    "starsDisplay": "9.9k",
    "weeklyStars": 1029,
    "weeklyGrowth": "+9.8%",
    "monthlyGrowth": "+39%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 39,
    "trafficNum": 890000,
    "searchVolNum": 140000,
    "searchTrend": "+98%",
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "elo": 1289,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/accelerate",
    "rank": 179
  },
  {
    "id": "astral-rye",
    "name": "Rye",
    "repo": "astral-sh/rye",
    "author": "astral-sh",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Python Systems Engineers",
    "stars": 14137,
    "wGrowth": 9.8,
    "mGrowth": 38,
    "website": "https://rye.astral.sh",
    "traffic": "480k/mo",
    "searchVol": "75,000/mo",
    "latency": 14,
    "tokens": 97,
    "desc": "An Experimental Package Management Solution for Python by Armin Ronacher, now maintained by Astral.",
    "useCase": "Manage Python toolchains and virtual environments deterministically in automated build pipelines.",
    "arch": "Rust, Cargo-like workflow, Global tool caching.",
    "tags": [
      "python-toolchain",
      "astral",
      "rust",
      "package-management"
    ],
    "starsDisplay": "14.1k",
    "weeklyStars": 1960,
    "weeklyGrowth": "+9.8%",
    "monthlyGrowth": "+38%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 38,
    "trafficNum": 480000,
    "searchVolNum": 75000,
    "searchTrend": "+98%",
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "elo": 1289,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/astral-sh.png",
    "githubUrl": "https://github.com/astral-sh/rye",
    "rank": 180
  },
  {
    "id": "pdfplumber",
    "name": "pdfplumber",
    "repo": "jsvine/pdfplumber",
    "author": "jsvine",
    "persona": "STEM & AI Researchers",
    "targetRole": "Visual PDF Extraction Devs",
    "stars": 10794,
    "wGrowth": 66.1,
    "mGrowth": 40,
    "website": "https://github.com/jsvine/pdfplumber",
    "traffic": "540k/mo",
    "searchVol": "98,000/mo",
    "latency": 12,
    "tokens": 97,
    "desc": "Plumb a PDF for detailed information about each text character, rectangle, and line, plus table extraction.",
    "useCase": "Extract pixel-precise layout data and nested tabular data from PDF statements and financial filings.",
    "arch": "Python, pdfminer.six, Pillow visual debugging.",
    "tags": [
      "pdf-inspection",
      "table-extractor",
      "character-coordinates",
      "python"
    ],
    "starsDisplay": "10.8k",
    "weeklyStars": 4931,
    "weeklyGrowth": "+66.1%",
    "monthlyGrowth": "+40%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 40,
    "trafficNum": 540000,
    "searchVolNum": 98000,
    "searchTrend": "+98%",
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "elo": 1289,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/jsvine.png",
    "githubUrl": "https://github.com/jsvine/pdfplumber",
    "rank": 181
  },
  {
    "id": "bark",
    "name": "Bark",
    "repo": "suno-ai/bark",
    "author": "suno-ai",
    "persona": "Music & Audio Creators",
    "targetRole": "Generative Audio Pioneer Leads",
    "stars": 39272,
    "wGrowth": 9.8,
    "mGrowth": 39,
    "website": "https://github.com/suno-ai/bark",
    "traffic": "1.3M/mo",
    "searchVol": "230,000/mo",
    "latency": 32,
    "tokens": 86,
    "desc": "Transformer-based audio generation model by Suno, creating highly realistic speech, music, and ambient noise.",
    "useCase": "Generate expressive voice tracks with human laughter, crying, sighing, and background musical accompaniment.",
    "arch": "Python, PyTorch, EnCodec, GPT-style autoregressive model.",
    "tags": [
      "suno",
      "generative-audio",
      "speech-synthesis",
      "bark"
    ],
    "starsDisplay": "39.3k",
    "weeklyStars": 6980,
    "weeklyGrowth": "+9.8%",
    "monthlyGrowth": "+39%",
    "wGrowthNum": 9.8,
    "mGrowthNum": 39,
    "trafficNum": 13000000,
    "searchVolNum": 230000,
    "searchTrend": "+98%",
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "elo": 1292,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/suno-ai.png",
    "githubUrl": "https://github.com/suno-ai/bark",
    "rank": 182
  },
  {
    "id": "yc-hackathon-stack",
    "name": "YC Hackathon Starter",
    "repo": "ycombinator/hackathon-stack",
    "author": "ycombinator",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Speed-Run Startup Builders",
    "stars": 16500,
    "wGrowth": "+9.6%",
    "mGrowth": "+38.4%",
    "website": "https://ycombinator.com",
    "traffic": "940k/mo",
    "trafficNum": 940000,
    "searchVol": "76,000/mo",
    "searchVolNum": 76000,
    "searchTrend": "+72% (Strong)",
    "latency": "25ms",
    "tokens": "-40%",
    "desc": "Production-ready full-stack template created for Y Combinator AI hackathons with Next.js App Router, Tailwind, Supabase, and Stripe pre-configured.",
    "useCase": "Going from idea to deployed revenue-collecting agentic SaaS in less than 24 hours with battle-tested YC conventions.",
    "arch": "Next.js App Router, TypeScript, Tailwind CSS, Supabase Auth/DB, Stripe Elements.",
    "tags": [
      "Y Combinator",
      "Hackathon",
      "Next.js",
      "Supabase",
      "Stripe"
    ],
    "wGrowthNum": 9.6,
    "mGrowthNum": 38.4,
    "latencyNum": 25,
    "tokensNum": 40,
    "weeklyStars": 1584,
    "weeklyGrowth": "+9.6%",
    "monthlyGrowth": "+38.4%",
    "latencyMs": "25ms",
    "tokensSaved": "-40%",
    "starsDisplay": "16.5k",
    "logoUrl": "https://github.com/ycombinator.png",
    "githubUrl": "https://github.com/ycombinator/hackathon-stack",
    "rank": 183,
    "elo": 1209,
    "eloCi": "±23",
    "winRate": 74
  },
  {
    "id": "sourcegraph-cody",
    "name": "Sourcegraph Cody",
    "repo": "sourcegraph/cody",
    "author": "sourcegraph",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Enterprise SWE Architects",
    "stars": 12800,
    "wGrowth": 9.6,
    "mGrowth": 38,
    "website": "https://sourcegraph.com/cody",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "175,000/mo",
    "searchVolNum": 175000,
    "searchTrend": "+130%",
    "latency": 24,
    "tokens": 88,
    "desc": "AI code assistant that writes code, fixes bugs, and explains code by leveraging Sourcegraph's code graph.",
    "useCase": "Answer architectural queries spanning multi-million line enterprise codebases with authoritative AST references.",
    "arch": "TypeScript, Sourcegraph Code Graph, Vector embeddings, VS Code / JetBrains extension.",
    "tags": [
      "enterprise-coding",
      "code-graph",
      "ast-search",
      "cody"
    ],
    "elo": 1305,
    "eloCi": "±20",
    "winRate": 86,
    "logoUrl": "https://github.com/sourcegraph.png",
    "githubUrl": "https://github.com/sourcegraph/cody",
    "starsDisplay": "12.8k",
    "weeklyStars": 1229,
    "weeklyGrowth": "+9.6%",
    "monthlyGrowth": "+38.0%",
    "wGrowthNum": 9.6,
    "mGrowthNum": 38,
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 184
  },
  {
    "id": "firecrawl",
    "name": "Firecrawl",
    "repo": "mendableai/firecrawl",
    "author": "mendableai",
    "persona": "AI Ops & Tooling",
    "targetRole": "Data Pipeline Engineers",
    "stars": 189290,
    "wGrowth": 768.3,
    "mGrowth": 41,
    "website": "https://firecrawl.dev",
    "traffic": "1.5M/mo",
    "trafficNum": 1500000,
    "searchVol": "72,000/mo",
    "searchVolNum": 72000,
    "searchTrend": "+98%",
    "latency": 45,
    "tokens": 94,
    "desc": "Turn entire websites into clean LLM-ready markdown. Handles proxies, anti-bot protection, dynamic JS rendering, and subpage crawling.",
    "useCase": "Convert an entire knowledge base or competitor documentation directory into structured JSON or Markdown with single API calls.",
    "arch": "TypeScript, BullMQ, Redis, Puppeteer cluster, Rust HTML sanitizers.",
    "tags": [
      "crawler",
      "markdown",
      "anti-bot",
      "web-intelligence"
    ],
    "rank": 185,
    "elo": 1219,
    "eloCi": "±24",
    "winRate": 76,
    "starsDisplay": "189.3k",
    "weeklyStars": 169561,
    "weeklyGrowth": "+768.3%",
    "monthlyGrowth": "+41.0%",
    "wGrowthNum": 9.5,
    "mGrowthNum": 41,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/mendableai.png",
    "githubUrl": "https://github.com/mendableai/firecrawl",
    "todayBreakthrough": "Fire"
  },
  {
    "id": "bun",
    "name": "Bun Runtime",
    "repo": "oven-sh/bun",
    "author": "oven-sh",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Full-Stack & Systems Engineers",
    "stars": 96141,
    "wGrowth": 25.7,
    "mGrowth": 36,
    "website": "https://bun.sh",
    "traffic": "3.5M/mo",
    "trafficNum": 3500000,
    "searchVol": "580,000/mo",
    "searchVolNum": 580000,
    "searchTrend": "+135%",
    "latency": 10,
    "tokens": 95,
    "desc": "Incredibly fast JavaScript & TypeScript runtime, bundler, test runner, and package manager. Built from scratch in Zig.",
    "useCase": "Accelerate agentic code generation runs, package installations, and test executions by 4x to 10x over Node.js.",
    "arch": "Zig, JavaScriptCore (JSC), POSIX syscall optimizations.",
    "tags": [
      "javascript-runtime",
      "zig",
      "package-manager",
      "high-performance"
    ],
    "elo": 1360,
    "eloCi": "±16",
    "winRate": 92,
    "logoUrl": "https://github.com/oven-sh.png",
    "githubUrl": "https://github.com/oven-sh/bun",
    "starsDisplay": "96.1k",
    "weeklyStars": 26909,
    "weeklyGrowth": "+25.7%",
    "monthlyGrowth": "+36.0%",
    "wGrowthNum": 9.5,
    "mGrowthNum": 36,
    "latencyMs": "10ms",
    "latencyNum": 10,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "rank": 186
  },
  {
    "id": "tokenizers",
    "name": "HuggingFace Tokenizers",
    "repo": "huggingface/tokenizers",
    "author": "huggingface",
    "persona": "Autonomous Software Engineers",
    "targetRole": "NLP Engineers",
    "stars": 11158,
    "wGrowth": 21.3,
    "mGrowth": 36.1,
    "website": "https://github.com/huggingface/tokenizers",
    "traffic": "460,000/mo",
    "trafficNum": 460000,
    "searchVol": "92,000/mo",
    "searchVolNum": 92000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool HuggingFace Tokenizers powering production AI workloads.",
    "useCase": "Integrate HuggingFace Tokenizers into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1318,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "11.2k",
    "weeklyStars": 2832,
    "weeklyGrowth": "+21.3%",
    "monthlyGrowth": "+36.1%",
    "wGrowthNum": 9.5,
    "mGrowthNum": 36.1,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/tokenizers",
    "rank": 187
  },
  {
    "id": "composio",
    "name": "Composio",
    "repo": "ComposioHQ/composio",
    "author": "ComposioHQ",
    "persona": "AI Ops & Tooling",
    "targetRole": "Integration Engineers",
    "stars": 30455,
    "wGrowth": 105.8,
    "mGrowth": 40.5,
    "website": "https://composio.dev",
    "traffic": "650k/mo",
    "trafficNum": 650000,
    "searchVol": "45,000/mo",
    "searchVolNum": 45000,
    "searchTrend": "+88%",
    "latency": 45,
    "tokens": 91,
    "desc": "250+ tools and actions for AI agents with managed authentication and OAuth.",
    "useCase": "Give Claude or agents instant access to GitHub PR creation, Gmail sending, and Slack notifications.",
    "arch": "Python, TypeScript, OAuth proxy.",
    "tags": [
      "tool-calling",
      "oauth",
      "actions"
    ],
    "rank": 188,
    "elo": 1204,
    "eloCi": "±25",
    "winRate": 74,
    "starsDisplay": "30.5k",
    "weeklyStars": 17046,
    "weeklyGrowth": "+105.8%",
    "monthlyGrowth": "+40.5%",
    "wGrowthNum": 9.4,
    "mGrowthNum": 40.5,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/ComposioHQ.png",
    "githubUrl": "https://github.com/ComposioHQ/composio"
  },
  {
    "id": "llama-cpp",
    "name": "llama.cpp",
    "repo": "ggerganov/llama.cpp",
    "author": "ggerganov",
    "persona": "Researchers & STEM",
    "targetRole": "Hardware & Edge AI Engineers",
    "stars": 130544,
    "wGrowth": 75.9,
    "mGrowth": 37,
    "website": "https://github.com/ggerganov/llama.cpp",
    "traffic": "4.5M/mo",
    "trafficNum": 4500000,
    "searchVol": "720,000/mo",
    "searchVolNum": 720000,
    "searchTrend": "+125%",
    "latency": 22,
    "tokens": 96,
    "desc": "Port of Facebook's LLaMA model in pure C/C++ with zero dependencies. Runs high-parameter models on consumer hardware.",
    "useCase": "Run 4-bit quantized frontier LLMs locally on Apple Silicon Metal, CUDA, and plain x86 CPU at maximum tokens/sec.",
    "arch": "C/C++, GGML / GGUF format, Apple Metal, AVX-512, CUDA.",
    "tags": [
      "local-llm",
      "gguf",
      "c-plus-plus",
      "apple-metal"
    ],
    "elo": 1365,
    "eloCi": "±15",
    "winRate": 93,
    "logoUrl": "https://github.com/ggerganov.png",
    "githubUrl": "https://github.com/ggerganov/llama.cpp",
    "starsDisplay": "130.5k",
    "weeklyStars": 63319,
    "weeklyGrowth": "+75.9%",
    "monthlyGrowth": "+37.0%",
    "wGrowthNum": 9.4,
    "mGrowthNum": 37,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "rank": 189
  },
  {
    "id": "tree-sitter",
    "name": "Tree-sitter",
    "repo": "tree-sitter/tree-sitter",
    "author": "tree-sitter",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Compiler & AST Architects",
    "stars": 27133,
    "wGrowth": 37,
    "mGrowth": 41,
    "website": "https://tree-sitter.github.io",
    "traffic": "780k/mo",
    "searchVol": "110,000/mo",
    "latency": 8,
    "tokens": 99,
    "desc": "Incremental parsing system for programming tools, providing syntax trees for code-analysis agents.",
    "useCase": "Parse ASTs incrementally in milliseconds so code-repair agents understand semantic AST structures without regex.",
    "arch": "C/Rust, Context-free grammars, Incremental diffing.",
    "tags": [
      "ast-parser",
      "syntax-tree",
      "language-tooling",
      "compilers"
    ],
    "starsDisplay": "27.1k",
    "weeklyStars": 9194,
    "weeklyGrowth": "+37%",
    "monthlyGrowth": "+41%",
    "wGrowthNum": 9.4,
    "mGrowthNum": 41,
    "trafficNum": 780000,
    "searchVolNum": 110000,
    "searchTrend": "+94%",
    "latencyMs": "8ms",
    "latencyNum": 8,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "elo": 1287,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/tree-sitter.png",
    "githubUrl": "https://github.com/tree-sitter/tree-sitter",
    "rank": 190
  },
  {
    "id": "cloudflare-browser-rendering",
    "name": "Cloudflare Browser Rendering",
    "repo": "cloudflare/puppeteer",
    "author": "cloudflare",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Headless Automation Engineers",
    "stars": 376,
    "wGrowth": "+9.2%",
    "mGrowth": "+34.1%",
    "website": "https://developers.cloudflare.com/browser-rendering",
    "traffic": "310k/mo",
    "trafficNum": 310000,
    "searchVol": "22,000/mo",
    "searchVolNum": 22000,
    "searchTrend": "+48% (Steady)",
    "latency": "180ms",
    "tokens": "-50%",
    "desc": "Headless browser automation running directly within Cloudflare Workers isolates for programmatic web scraping and screenshot generation.",
    "useCase": "Executing resilient browser automation and HTML parsing without managing dedicated Chromium container fleets.",
    "arch": "Puppeteer API, Cloudflare Workers, Sandbox isolation, Edge DOM.",
    "tags": [
      "Browser Automation",
      "Puppeteer",
      "Web Scraping",
      "Edge DOM",
      "Cloudflare"
    ],
    "wGrowthNum": 9.2,
    "mGrowthNum": 34.1,
    "latencyNum": 180,
    "tokensNum": 50,
    "weeklyStars": 901,
    "weeklyGrowth": "+9.2%",
    "monthlyGrowth": "+34.1%",
    "latencyMs": "180ms",
    "tokensSaved": "-50%",
    "starsDisplay": "376",
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/puppeteer",
    "rank": 191,
    "elo": 1192,
    "eloCi": "±26",
    "winRate": 72
  },
  {
    "id": "gpt-pilot",
    "name": "GPT-Pilot",
    "repo": "Pythagora-io/gpt-pilot",
    "author": "Pythagora-io",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Solo Founders & Builders",
    "stars": 33655,
    "wGrowth": 9.2,
    "mGrowth": 36,
    "website": "https://pythagora.ai",
    "traffic": "820k/mo",
    "trafficNum": 820000,
    "searchVol": "190,000/mo",
    "searchVolNum": 190000,
    "searchTrend": "+105%",
    "latency": 45,
    "tokens": 76,
    "desc": "Dev tool that writes code step by step while the developer reviews each file, test, and command before proceeding.",
    "useCase": "Build complete end-to-end full stack web applications from a product brief with automated step-by-step verification.",
    "arch": "Python, Electron GUI, SQLite, OS Shell isolation.",
    "tags": [
      "full-stack-builder",
      "developer-agent",
      "step-by-step",
      "code-generator"
    ],
    "elo": 1275,
    "eloCi": "±22",
    "winRate": 82,
    "logoUrl": "https://github.com/Pythagora-io.png",
    "githubUrl": "https://github.com/Pythagora-io/gpt-pilot",
    "starsDisplay": "33.7k",
    "weeklyStars": 5325,
    "weeklyGrowth": "+9.2%",
    "monthlyGrowth": "+36.0%",
    "wGrowthNum": 9.2,
    "mGrowthNum": 36,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-76%",
    "tokensNum": 76,
    "rank": 192
  },
  {
    "id": "mem0",
    "name": "Mem0",
    "repo": "mem0ai/mem0",
    "author": "mem0ai",
    "persona": "AI Ops & Tooling",
    "targetRole": "Agent Architecture Leads",
    "stars": 66738,
    "wGrowth": 162.7,
    "mGrowth": 39.5,
    "website": "https://mem0.ai",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "78,000/mo",
    "searchVolNum": 78000,
    "searchTrend": "+85%",
    "latency": 65,
    "tokens": 92,
    "desc": "The Memory Layer for AI Agents. Enables continuous learning, personalized user memory, and hierarchical context across sessions.",
    "useCase": "Store user preferences, past tool outcomes, and historical constraints so the AI remembers past lessons on subsequent sessions.",
    "arch": "Hybrid Graph-Vector store, Qdrant/pgvector, Python client.",
    "tags": [
      "memory-layer",
      "agent-state",
      "personalization",
      "vector-graph"
    ],
    "rank": 193,
    "elo": 1222,
    "eloCi": "±18",
    "winRate": 76,
    "starsDisplay": "66.7k",
    "weeklyStars": 43649,
    "weeklyGrowth": "+162.7%",
    "monthlyGrowth": "+39.5%",
    "wGrowthNum": 9.1,
    "mGrowthNum": 39.5,
    "latencyMs": "65ms",
    "latencyNum": 65,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/mem0ai.png",
    "githubUrl": "https://github.com/mem0ai/mem0"
  },
  {
    "id": "stagehand",
    "name": "Stagehand",
    "repo": "browserbase/stagehand",
    "author": "browserbase",
    "persona": "AI Ops & Tooling",
    "targetRole": "Browser Automation Devs",
    "stars": 25556,
    "wGrowth": 223.5,
    "mGrowth": 39,
    "website": "https://stagehand.dev",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "28,000/mo",
    "searchVolNum": 28000,
    "searchTrend": "+88%",
    "latency": 55,
    "tokens": 93,
    "desc": "An open source browser automation framework for AI agents built on Playwright.",
    "useCase": "Automate complex web interactions using intent-based act and observe methods.",
    "arch": "Playwright, TypeScript, AI DOM extractor.",
    "tags": [
      "playwright",
      "browser-agent",
      "intent"
    ],
    "rank": 194,
    "elo": 1187,
    "eloCi": "±19",
    "winRate": 71,
    "starsDisplay": "25.6k",
    "weeklyStars": 18374,
    "weeklyGrowth": "+223.5%",
    "monthlyGrowth": "+39.0%",
    "wGrowthNum": 9.1,
    "mGrowthNum": 39,
    "latencyMs": "55ms",
    "latencyNum": 55,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/browserbase.png",
    "githubUrl": "https://github.com/browserbase/stagehand"
  },
  {
    "id": "weaviate",
    "name": "Weaviate",
    "repo": "weaviate/weaviate",
    "author": "weaviate",
    "persona": "AI Ops & Tooling",
    "targetRole": "MLOps & Search Architects",
    "stars": 16870,
    "wGrowth": 30.8,
    "mGrowth": 35,
    "website": "https://weaviate.io",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "165,000/mo",
    "searchVolNum": 165000,
    "searchTrend": "+120%",
    "latency": 20,
    "tokens": 92,
    "desc": "Open-source vector database that allows developers to store data objects and vector embeddings from ML models.",
    "useCase": "Perform hybrid search combining BM25 keyword matching with vector distance at enterprise scale.",
    "arch": "Go, Custom HNSW index, GraphQL API, Raft consensus.",
    "tags": [
      "vector-database",
      "hybrid-search",
      "go",
      "enterprise-search"
    ],
    "elo": 1310,
    "eloCi": "±19",
    "winRate": 86,
    "logoUrl": "https://github.com/weaviate.png",
    "githubUrl": "https://github.com/weaviate/weaviate",
    "starsDisplay": "16.9k",
    "weeklyStars": 5144,
    "weeklyGrowth": "+30.8%",
    "monthlyGrowth": "+35.0%",
    "wGrowthNum": 9.1,
    "mGrowthNum": 35,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 195
  },
  {
    "id": "onnxruntime",
    "name": "ONNX Runtime",
    "repo": "microsoft/onnxruntime",
    "author": "microsoft",
    "persona": "Cloudflare & Edge Architects",
    "targetRole": "Cross-Platform ML Engineers",
    "stars": 22029,
    "wGrowth": 28.1,
    "mGrowth": 36,
    "website": "https://onnxruntime.ai",
    "traffic": "1.4M/mo",
    "searchVol": "220,000/mo",
    "latency": 12,
    "tokens": 98,
    "desc": "Cross-platform, high-performance ML inferencing and training accelerator supporting CUDA, TensorRT, CoreML.",
    "useCase": "Deploy neural models across Linux servers, Apple Silicon, browser WASM, and mobile devices.",
    "arch": "C++, ONNX graph optimizer, Execution Providers (EP).",
    "tags": [
      "onnx",
      "cross-platform",
      "edge-inference",
      "microsoft"
    ],
    "starsDisplay": "22.0k",
    "weeklyStars": 6394,
    "weeklyGrowth": "+28.1%",
    "monthlyGrowth": "+36%",
    "wGrowthNum": 9.1,
    "mGrowthNum": 36,
    "trafficNum": 1000000,
    "searchVolNum": 220000,
    "searchTrend": "+91%",
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "elo": 1285,
    "eloCi": "±18",
    "winRate": 84,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/onnxruntime",
    "rank": 196
  },
  {
    "id": "zod",
    "name": "Zod Schema",
    "repo": "colinhacks/zod",
    "author": "colinhacks",
    "persona": "Founders & Indie Hackers",
    "targetRole": "TypeScript Engineers",
    "stars": 44066,
    "wGrowth": 26.6,
    "mGrowth": 34.2,
    "website": "https://github.com/colinhacks/zod",
    "traffic": "1,740,000/mo",
    "trafficNum": 1740000,
    "searchVol": "348,000/mo",
    "searchVolNum": 348000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Zod Schema powering production AI workloads.",
    "useCase": "Integrate Zod Schema into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1316,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "44.1k",
    "weeklyStars": 12398,
    "weeklyGrowth": "+26.6%",
    "monthlyGrowth": "+34.2%",
    "wGrowthNum": 9,
    "mGrowthNum": 34.2,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/colinhacks.png",
    "githubUrl": "https://github.com/colinhacks/zod",
    "rank": 197
  },
  {
    "id": "marimo",
    "name": "Marimo",
    "repo": "marimo-team/marimo",
    "author": "marimo-team",
    "persona": "Researchers & STEM",
    "targetRole": "Computational Scientists",
    "stars": 23044,
    "wGrowth": 135.1,
    "mGrowth": 38.5,
    "website": "https://marimo.io",
    "traffic": "520k/mo",
    "trafficNum": 520000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+88%",
    "latency": 32,
    "tokens": 95,
    "desc": "A reactive Python notebook for data science and machine learning. Reproducible by design.",
    "useCase": "Build interactive computational notebooks with deterministic DAG execution.",
    "arch": "Python, WebSockets, Reactive DAG.",
    "tags": [
      "notebook",
      "reactive",
      "reproducible-science"
    ],
    "rank": 198,
    "elo": 1190,
    "eloCi": "±20",
    "winRate": 72,
    "starsDisplay": "23.0k",
    "weeklyStars": 14116,
    "weeklyGrowth": "+135.1%",
    "monthlyGrowth": "+38.5%",
    "wGrowthNum": 8.9,
    "mGrowthNum": 38.5,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/marimo-team.png",
    "githubUrl": "https://github.com/marimo-team/marimo"
  },
  {
    "id": "cloudflare-workers-ai",
    "name": "Cloudflare Workers AI",
    "repo": "cloudflare/workers-ai",
    "author": "cloudflare",
    "persona": "AI Ops & Tooling",
    "targetRole": "Serverless AI Architects",
    "stars": 22800,
    "wGrowth": "+8.9%",
    "mGrowth": "+36.5%",
    "website": "https://ai.cloudflare.com",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "84,000/mo",
    "searchVolNum": 84000,
    "searchTrend": "+62% (Strong)",
    "latency": "45ms",
    "tokens": "-38%",
    "desc": "Serverless GPU inference running open models (Llama 3.3, DeepSeek-R1, Whisper, Mistral) directly on Cloudflare global edge network.",
    "useCase": "Running production LLM inference and embeddings without managing GPU instances, clusters, or auto-scaling groups.",
    "arch": "Cloudflare Edge GPU mesh, V8 isolates, Zero-cold-start inference.",
    "tags": [
      "Serverless GPU",
      "Llama 3.3",
      "DeepSeek",
      "Edge Inference",
      "Cloudflare"
    ],
    "wGrowthNum": 8.9,
    "mGrowthNum": 36.5,
    "latencyNum": 45,
    "tokensNum": 38,
    "weeklyStars": 2029,
    "weeklyGrowth": "+8.9%",
    "monthlyGrowth": "+36.5%",
    "latencyMs": "45ms",
    "tokensSaved": "-38%",
    "starsDisplay": "22.8k",
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/workers-ai",
    "rank": 199,
    "elo": 1216,
    "eloCi": "±21",
    "winRate": 75
  },
  {
    "id": "tabby",
    "name": "TabbyML",
    "repo": "TabbyML/tabby",
    "author": "TabbyML",
    "persona": "AI Ops & Tooling",
    "targetRole": "Enterprise Infrastructure & Security",
    "stars": 33902,
    "wGrowth": 27.9,
    "mGrowth": 35,
    "website": "https://tabby.tabbyml.com",
    "traffic": "680k/mo",
    "trafficNum": 680000,
    "searchVol": "135,000/mo",
    "searchVolNum": 135000,
    "searchTrend": "+115%",
    "latency": 19,
    "tokens": 90,
    "desc": "Self-hosted AI coding assistant. An open-source, on-premise alternative to GitHub Copilot with zero telemetry.",
    "useCase": "Deploy private code completion and chat models inside sovereign VPCs with zero data exfiltration.",
    "arch": "Rust, llama.cpp / vLLM backend, Tree-sitter, Triton server.",
    "tags": [
      "self-hosted",
      "copilot-alternative",
      "rust",
      "sovereign-ai"
    ],
    "elo": 1315,
    "eloCi": "±19",
    "winRate": 87,
    "logoUrl": "https://github.com/TabbyML.png",
    "githubUrl": "https://github.com/TabbyML/tabby",
    "starsDisplay": "33.9k",
    "weeklyStars": 9761,
    "weeklyGrowth": "+27.9%",
    "monthlyGrowth": "+35.0%",
    "wGrowthNum": 8.9,
    "mGrowthNum": 35,
    "latencyMs": "19ms",
    "latencyNum": 19,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "rank": 200
  },
  {
    "id": "twenty-crm",
    "name": "Twenty CRM",
    "repo": "twentyhq/twenty",
    "author": "twentyhq",
    "persona": "Founders & Indie Hackers",
    "targetRole": "B2B Founders & Sales",
    "stars": 58014,
    "wGrowth": 133.9,
    "mGrowth": 38.5,
    "website": "https://twenty.com",
    "traffic": "620k/mo",
    "trafficNum": 620000,
    "searchVol": "38,000/mo",
    "searchVolNum": 38000,
    "searchTrend": "+85%",
    "latency": 52,
    "tokens": 87,
    "desc": "A modern alternative to Salesforce. Flexible, powerful, community-driven.",
    "useCase": "Manage venture capital outreach, B2B enterprise deals, and custom client pipelines.",
    "arch": "TypeScript, NestJS, React, PostgreSQL.",
    "tags": [
      "crm",
      "salesforce-alt",
      "pipeline"
    ],
    "rank": 201,
    "elo": 1219,
    "eloCi": "±22",
    "winRate": 76,
    "starsDisplay": "58.0k",
    "weeklyStars": 35396,
    "weeklyGrowth": "+133.9%",
    "monthlyGrowth": "+38.5%",
    "wGrowthNum": 8.8,
    "mGrowthNum": 38.5,
    "latencyMs": "52ms",
    "latencyNum": 52,
    "tokensSaved": "-87%",
    "tokensNum": 87,
    "logoUrl": "https://github.com/twentyhq.png",
    "githubUrl": "https://github.com/twentyhq/twenty"
  },
  {
    "id": "zed-editor",
    "name": "Zed Editor",
    "repo": "zed-industries/zed",
    "author": "zed-industries",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Systems Engineers & High-Performance Devs",
    "stars": 91382,
    "wGrowth": 74.4,
    "mGrowth": 34,
    "website": "https://zed.dev",
    "traffic": "2.6M/mo",
    "trafficNum": 2600000,
    "searchVol": "380,000/mo",
    "searchVolNum": 380000,
    "searchTrend": "+135%",
    "latency": 12,
    "tokens": 94,
    "desc": "High-performance, multiplayer code editor written in Rust with built-in native LLM chat, completions, and model transformation.",
    "useCase": "Zero-latency native coding with streaming Claude 3.5 and inline model-assisted transformations.",
    "arch": "Rust, GPUI hardware-accelerated UI, Tree-sitter, CRDT multiplayer.",
    "tags": [
      "code-editor",
      "rust",
      "native-performance",
      "gpui"
    ],
    "elo": 1350,
    "eloCi": "±16",
    "winRate": 91,
    "logoUrl": "https://github.com/zed-industries.png",
    "githubUrl": "https://github.com/zed-industries/zed",
    "starsDisplay": "91.4k",
    "weeklyStars": 43593,
    "weeklyGrowth": "+74.4%",
    "monthlyGrowth": "+34.0%",
    "wGrowthNum": 8.8,
    "mGrowthNum": 34,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "rank": 202
  },
  {
    "id": "uv",
    "name": "uv",
    "repo": "astral-sh/uv",
    "author": "astral-sh",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Python Environment Engineers",
    "stars": 90461,
    "wGrowth": 116.4,
    "mGrowth": 37,
    "website": "https://astral.sh/uv",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "110,000/mo",
    "searchVolNum": 110000,
    "searchTrend": "+90%",
    "latency": 12,
    "tokens": 99,
    "desc": "An extremely fast Python package and project manager, written in Rust.",
    "useCase": "Install complex agent dependencies and setup virtual environments in 50ms.",
    "arch": "Rust, PubGrub resolution algorithm.",
    "tags": [
      "package-manager",
      "pip-alt",
      "virtualenv"
    ],
    "rank": 203,
    "elo": 1251,
    "eloCi": "±23",
    "winRate": 80,
    "starsDisplay": "90.5k",
    "weeklyStars": 52255,
    "weeklyGrowth": "+116.4%",
    "monthlyGrowth": "+37.0%",
    "wGrowthNum": 8.6,
    "mGrowthNum": 37,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "logoUrl": "https://github.com/astral-sh.png",
    "githubUrl": "https://github.com/astral-sh/uv"
  },
  {
    "id": "framer-motion",
    "name": "Framer Motion",
    "repo": "framer/motion",
    "author": "framer",
    "persona": "Founders & Indie Hackers",
    "targetRole": "UI/UX & Product Designers",
    "stars": 33850,
    "wGrowth": 26.3,
    "mGrowth": 33,
    "website": "https://framer.com/motion",
    "traffic": "1.9M/mo",
    "trafficNum": 1900000,
    "searchVol": "310,000/mo",
    "searchVolNum": 310000,
    "searchTrend": "+110%",
    "latency": 16,
    "tokens": 92,
    "desc": "Production-ready motion library for React. Declarative syntax, gesture handling, and physics-based spring simulations.",
    "useCase": "Deliver smooth drawer slides, page transitions, and interactive visual indicators without CSS keyframe complexity.",
    "arch": "TypeScript, Web Animations API, Popmotion physics engine.",
    "tags": [
      "animation",
      "react",
      "motion",
      "ui-ux"
    ],
    "elo": 1305,
    "eloCi": "±19",
    "winRate": 86,
    "logoUrl": "https://github.com/framer.png",
    "githubUrl": "https://github.com/framer/motion",
    "starsDisplay": "33.9k",
    "weeklyStars": 9355,
    "weeklyGrowth": "+26.3%",
    "monthlyGrowth": "+33.0%",
    "wGrowthNum": 8.6,
    "mGrowthNum": 33,
    "latencyMs": "16ms",
    "latencyNum": 16,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 204
  },
  {
    "id": "gpt4all",
    "name": "GPT4All",
    "repo": "nomic-ai/gpt4all",
    "author": "nomic-ai",
    "persona": "Founders & Solo Creators",
    "targetRole": "Edge Inference Developers",
    "stars": 77380,
    "wGrowth": 8.6,
    "mGrowth": 34,
    "website": "https://gpt4all.io",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "280,000/mo",
    "searchVolNum": 280000,
    "searchTrend": "+85%",
    "latency": 34,
    "tokens": 90,
    "desc": "Open-source ecosystem of consumer hardware-optimized LLMs and desktop applications for private on-device intelligence.",
    "useCase": "Chat with personal local documents and code repositories without internet access or GPU clusters.",
    "arch": "C/C++, Qt desktop, Vulkan/CUDA acceleration, ggml/llama.cpp.",
    "tags": [
      "local-llm",
      "desktop-ai",
      "local-docs",
      "nomic"
    ],
    "elo": 1300,
    "eloCi": "±18",
    "winRate": 86,
    "starsDisplay": "77.4k",
    "weeklyStars": 12029,
    "weeklyGrowth": "+8.6%",
    "monthlyGrowth": "+34%",
    "wGrowthNum": 8.6,
    "mGrowthNum": 34,
    "latencyMs": "34ms",
    "latencyNum": 34,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/nomic-ai.png",
    "githubUrl": "https://github.com/nomic-ai/gpt4all",
    "rank": 205
  },
  {
    "id": "chattts",
    "name": "ChatTTS",
    "repo": "2noise/ChatTTS",
    "author": "2noise",
    "persona": "Music & Audio Creators",
    "targetRole": "Dialogue Producers",
    "stars": 39888,
    "wGrowth": 23.1,
    "mGrowth": 36.8,
    "website": "https://chattts.com",
    "traffic": "890k/mo",
    "trafficNum": 890000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+92%",
    "latency": 65,
    "tokens": 92,
    "desc": "Conversational text-to-speech model optimized for interactive dialogue and dialogue prosody.",
    "useCase": "Synthesize natural dialogue pauses and conversational voiceovers.",
    "arch": "Autoregressive speech transformer, PyTorch.",
    "tags": [
      "conversational-tts",
      "natural-prosody",
      "dialogue"
    ],
    "rank": 206,
    "elo": 1232,
    "eloCi": "±24",
    "winRate": 78,
    "starsDisplay": "39.9k",
    "weeklyStars": 10242,
    "weeklyGrowth": "+23.1%",
    "monthlyGrowth": "+36.8%",
    "wGrowthNum": 8.5,
    "mGrowthNum": 36.8,
    "latencyMs": "65ms",
    "latencyNum": 65,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/2noise.png",
    "githubUrl": "https://github.com/2noise/ChatTTS"
  },
  {
    "id": "meta-llama-stack",
    "name": "Meta Llama Stack",
    "repo": "meta-llama/llama-stack",
    "author": "meta-llama",
    "persona": "AI Ops & Tooling",
    "targetRole": "Open-Source AI Infrastructure Engineers",
    "stars": 8435,
    "wGrowth": "+8.5%",
    "mGrowth": "+33.0%",
    "website": "https://llama.meta.com",
    "traffic": "780k/mo",
    "trafficNum": 780000,
    "searchVol": "58,000/mo",
    "searchVolNum": 58000,
    "searchTrend": "+64% (Strong)",
    "latency": "35ms",
    "tokens": "-35%",
    "desc": "Standardized APIs and agentic building blocks for evaluating, fine-tuning, and serving Llama models across cloud and on-premise hardware.",
    "useCase": "Building unified enterprise AI agent architectures that remain vendor-independent and fully sovereign on open weights.",
    "arch": "Python, PyTorch, Llama CLI, REST API specification, Ollama/vLLM backend.",
    "tags": [
      "Llama 3.3",
      "Open Weights",
      "Inference Stack",
      "Meta",
      "Standardized API"
    ],
    "wGrowthNum": 8.5,
    "mGrowthNum": 33,
    "latencyNum": 35,
    "tokensNum": 35,
    "weeklyStars": 1011,
    "weeklyGrowth": "+8.5%",
    "monthlyGrowth": "+33.0%",
    "latencyMs": "35ms",
    "tokensSaved": "-35%",
    "starsDisplay": "8.4k",
    "logoUrl": "https://github.com/meta-llama.png",
    "githubUrl": "https://github.com/meta-llama/llama-stack",
    "rank": 207,
    "elo": 1191,
    "eloCi": "±25",
    "winRate": 72
  },
  {
    "id": "llamaindex",
    "name": "LlamaIndex",
    "repo": "run-llama/llama_index",
    "author": "run-llama",
    "persona": "Autonomous Software Engineers",
    "targetRole": "RAG & Data Engineers",
    "stars": 52428,
    "wGrowth": 32.7,
    "mGrowth": 32,
    "website": "https://llamaindex.ai",
    "traffic": "4.1M/mo",
    "trafficNum": 4100000,
    "searchVol": "450,000/mo",
    "searchVolNum": 450000,
    "searchTrend": "+110%",
    "latency": 30,
    "tokens": 85,
    "desc": "Data framework for building LLM applications over external enterprise data through ingestion, indexing, and querying.",
    "useCase": "Connect private documents, relational databases, and APIs into agentic query engines and multi-document synthesis workflows.",
    "arch": "Python / TypeScript, Vector store abstraction, Document loaders, Evaluation framework.",
    "tags": [
      "rag",
      "data-framework",
      "vector-index",
      "enterprise-search"
    ],
    "elo": 1325,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/run-llama.png",
    "githubUrl": "https://github.com/run-llama/llama_index",
    "starsDisplay": "52.4k",
    "weeklyStars": 16286,
    "weeklyGrowth": "+32.7%",
    "monthlyGrowth": "+32.0%",
    "wGrowthNum": 8.5,
    "mGrowthNum": 32,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-85%",
    "tokensNum": 85,
    "rank": 208
  },
  {
    "id": "trpc",
    "name": "tRPC",
    "repo": "trpc/trpc",
    "author": "trpc",
    "persona": "Founders & Indie Hackers",
    "targetRole": "TypeScript Engineers",
    "stars": 40691,
    "wGrowth": 12.4,
    "mGrowth": 32.3,
    "website": "https://github.com/trpc/trpc",
    "traffic": "1,810,000/mo",
    "trafficNum": 1810000,
    "searchVol": "362,000/mo",
    "searchVolNum": 362000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool tRPC powering production AI workloads.",
    "useCase": "Integrate tRPC into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1314,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "40.7k",
    "weeklyStars": 7568,
    "weeklyGrowth": "+12.4%",
    "monthlyGrowth": "+32.3%",
    "wGrowthNum": 8.5,
    "mGrowthNum": 32.3,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/trpc.png",
    "githubUrl": "https://github.com/trpc/trpc",
    "rank": 209
  },
  {
    "id": "deepspeed",
    "name": "Microsoft DeepSpeed",
    "repo": "microsoft/DeepSpeed",
    "author": "microsoft",
    "persona": "STEM & AI Researchers",
    "targetRole": "Distributed Training Engineers",
    "stars": 43202,
    "wGrowth": 17.4,
    "mGrowth": 32,
    "website": "https://deepspeed.ai",
    "traffic": "780k/mo",
    "searchVol": "130,000/mo",
    "latency": 15,
    "tokens": 96,
    "desc": "Deep learning optimization software suite for extreme-scale distributed training and inference.",
    "useCase": "Train multi-billion parameter foundation models with ZeRO memory optimization across clusters.",
    "arch": "C++/CUDA, PyTorch, ZeRO memory stages.",
    "tags": [
      "distributed-training",
      "zero-memory",
      "microsoft",
      "gpu-optimization"
    ],
    "starsDisplay": "43.2k",
    "weeklyStars": 9530,
    "weeklyGrowth": "+17.4%",
    "monthlyGrowth": "+32%",
    "wGrowthNum": 8.5,
    "mGrowthNum": 32,
    "trafficNum": 780000,
    "searchVolNum": 130000,
    "searchTrend": "+85%",
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "elo": 1283,
    "eloCi": "±18",
    "winRate": 83,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/DeepSpeed",
    "rank": 210
  },
  {
    "id": "temporal",
    "name": "Temporal",
    "repo": "temporalio/temporal",
    "author": "temporalio",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Backend & Distributed Systems Engineers",
    "stars": 23510,
    "wGrowth": 78.1,
    "mGrowth": 32,
    "website": "https://temporal.io",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "165,000/mo",
    "searchVolNum": 165000,
    "searchTrend": "+115%",
    "latency": 20,
    "tokens": 92,
    "desc": "Durable execution platform that makes distributed applications failure-oblivious with code-first workflows.",
    "useCase": "Run long-running multi-day agentic transactions, payments, and data syncs without state corruption on restarts.",
    "arch": "Go, Event sourcing, Cassandra / PostgreSQL / MySQL persistence.",
    "tags": [
      "durable-execution",
      "workflow-engine",
      "distributed-systems",
      "fault-tolerant"
    ],
    "elo": 1345,
    "eloCi": "±17",
    "winRate": 90,
    "logoUrl": "https://github.com/temporalio.png",
    "githubUrl": "https://github.com/temporalio/temporal",
    "starsDisplay": "23.5k",
    "weeklyStars": 11419,
    "weeklyGrowth": "+78.1%",
    "monthlyGrowth": "+32.0%",
    "wGrowthNum": 8.4,
    "mGrowthNum": 32,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "rank": 211
  },
  {
    "id": "fooocus",
    "name": "Fooocus",
    "repo": "lllyasviel/Fooocus",
    "author": "lllyasviel",
    "persona": "Music & Audio Creators",
    "targetRole": "Visual Creators",
    "stars": 53314,
    "wGrowth": 26.6,
    "mGrowth": 31.9,
    "website": "https://github.com/lllyasviel/Fooocus",
    "traffic": "2,105,000/mo",
    "trafficNum": 2105000,
    "searchVol": "421,000/mo",
    "searchVolNum": 421000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Fooocus powering production AI workloads.",
    "useCase": "Integrate Fooocus into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1314,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "53.3k",
    "weeklyStars": 14750,
    "weeklyGrowth": "+26.6%",
    "monthlyGrowth": "+31.9%",
    "wGrowthNum": 8.4,
    "mGrowthNum": 31.9,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/lllyasviel.png",
    "githubUrl": "https://github.com/lllyasviel/Fooocus",
    "rank": 212
  },
  {
    "id": "turborepo",
    "name": "Turborepo",
    "repo": "vercel/turborepo",
    "author": "vercel",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Monorepo Devs",
    "stars": 31178,
    "wGrowth": 18.1,
    "mGrowth": 31.2,
    "website": "https://github.com/vercel/turborepo",
    "traffic": "1,320,000/mo",
    "trafficNum": 1320000,
    "searchVol": "264,000/mo",
    "searchVolNum": 264000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Turborepo powering production AI workloads.",
    "useCase": "Integrate Turborepo into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1313,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "31.2k",
    "weeklyStars": 6943,
    "weeklyGrowth": "+18.1%",
    "monthlyGrowth": "+31.2%",
    "wGrowthNum": 8.2,
    "mGrowthNum": 31.2,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/vercel.png",
    "githubUrl": "https://github.com/vercel/turborepo",
    "rank": 213
  },
  {
    "id": "tortoise-tts",
    "name": "Tortoise TTS",
    "repo": "neonbjb/tortoise-tts",
    "author": "neonbjb",
    "persona": "Music & Audio Creators",
    "targetRole": "High-Fidelity Audio Creators",
    "stars": 14875,
    "wGrowth": 20,
    "mGrowth": 32,
    "website": "https://github.com/neonbjb/tortoise-tts",
    "traffic": "450k/mo",
    "searchVol": "75,000/mo",
    "latency": 95,
    "tokens": 82,
    "desc": "Multi-voice text-to-speech system trained with an emphasis on quality, expressiveness and voice cloning.",
    "useCase": "Synthesize studio audiobook narration and high-realism voice acting.",
    "arch": "PyTorch, Autoregressive decoder, Diffusion acoustic model.",
    "tags": [
      "audiobooks",
      "voice-cloning",
      "expressive-speech",
      "diffusion-tts"
    ],
    "starsDisplay": "14.9k",
    "weeklyStars": 3492,
    "weeklyGrowth": "+20%",
    "monthlyGrowth": "+32%",
    "wGrowthNum": 8.2,
    "mGrowthNum": 32,
    "trafficNum": 450000,
    "searchVolNum": 75000,
    "searchTrend": "+82%",
    "latencyMs": "95ms",
    "latencyNum": 95,
    "tokensSaved": "-82%",
    "tokensNum": 82,
    "elo": 1278,
    "eloCi": "±18",
    "winRate": 83,
    "logoUrl": "https://github.com/neonbjb.png",
    "githubUrl": "https://github.com/neonbjb/tortoise-tts",
    "rank": 214
  },
  {
    "id": "scrapling",
    "name": "Scrapling",
    "repo": "D4Vinci/Scrapling",
    "author": "D4Vinci",
    "persona": "AI Ops & Tooling",
    "targetRole": "Stealth Scraping Leads",
    "stars": 86040,
    "wGrowth": 1812,
    "mGrowth": 35,
    "website": "https://github.com/D4Vinci/Scrapling",
    "traffic": "280k/mo",
    "trafficNum": 280000,
    "searchVol": "18,000/mo",
    "searchVolNum": 18000,
    "searchTrend": "+82%",
    "latency": 32,
    "tokens": 96,
    "desc": "Undetectable web scraping library with automated fingerprint spoofing and Camoufox engine.",
    "useCase": "Bypass Cloudflare Turnstile and Datadome bot detection with realistic browser signatures.",
    "arch": "Python, Camoufox, Camouflage headers.",
    "tags": [
      "stealth",
      "turnstile-bypass",
      "anti-bot"
    ],
    "rank": 215,
    "elo": 1173,
    "eloCi": "±26",
    "winRate": 69,
    "starsDisplay": "86.0k",
    "weeklyStars": 81904,
    "weeklyGrowth": "+1812%",
    "monthlyGrowth": "+35.0%",
    "wGrowthNum": 8.1,
    "mGrowthNum": 35,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/D4Vinci.png",
    "githubUrl": "https://github.com/D4Vinci/Scrapling"
  },
  {
    "id": "activepieces",
    "name": "Activepieces",
    "repo": "activepieces/activepieces",
    "author": "activepieces",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Growth & Automation Founders",
    "stars": 24930,
    "wGrowth": 94.8,
    "mGrowth": 34,
    "website": "https://activepieces.com",
    "traffic": "640k/mo",
    "trafficNum": 640000,
    "searchVol": "28,000/mo",
    "searchVolNum": 28000,
    "searchTrend": "+75%",
    "latency": 52,
    "tokens": 86,
    "desc": "Open-source business automation platform built for indie hackers and teams. Clean Zapier alternative with 100+ native pieces.",
    "useCase": "Connect Stripe checkout events to Slack notifications, Resend onboarding emails, and Google Sheets CRM without writing glue code.",
    "arch": "Node.js, PostgreSQL, Redis, Sandboxed V8 runtime, Tailwind.",
    "tags": [
      "automation",
      "zapier-alt",
      "webhooks",
      "ops"
    ],
    "rank": 216,
    "elo": 1188,
    "eloCi": "±18",
    "winRate": 71,
    "starsDisplay": "24.9k",
    "weeklyStars": 13141,
    "weeklyGrowth": "+94.8%",
    "monthlyGrowth": "+34.0%",
    "wGrowthNum": 7.9,
    "mGrowthNum": 34,
    "latencyMs": "52ms",
    "latencyNum": 52,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "logoUrl": "https://github.com/activepieces.png",
    "githubUrl": "https://github.com/activepieces/activepieces"
  },
  {
    "id": "tailwind-v4",
    "name": "Tailwind CSS",
    "repo": "tailwindlabs/tailwindcss",
    "author": "tailwindlabs",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Frontend & Design Engineers",
    "stars": 97796,
    "wGrowth": 17.3,
    "mGrowth": 30,
    "website": "https://tailwindcss.com",
    "traffic": "9.2M/mo",
    "trafficNum": 9200000,
    "searchVol": "1,200,000/mo",
    "searchVolNum": 1200000,
    "searchTrend": "+115%",
    "latency": 12,
    "tokens": 95,
    "desc": "Utility-first CSS framework for rapid UI development. v4 features oxide Rust engine for 10x faster build speeds.",
    "useCase": "Design clean, high-contrast, token-efficient responsive user interfaces directly in HTML without CSS bloat.",
    "arch": "Rust Oxide engine, Lightning CSS, Modern CSS nesting.",
    "tags": [
      "css-framework",
      "utility-first",
      "rust-compiler",
      "frontend"
    ],
    "elo": 1350,
    "eloCi": "±16",
    "winRate": 91,
    "logoUrl": "https://github.com/tailwindlabs.png",
    "githubUrl": "https://github.com/tailwindlabs/tailwindcss",
    "starsDisplay": "97.8k",
    "weeklyStars": 20985,
    "weeklyGrowth": "+17.3%",
    "monthlyGrowth": "+30.0%",
    "wGrowthNum": 7.9,
    "mGrowthNum": 30,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "rank": 217
  },
  {
    "id": "langfuse",
    "name": "Langfuse",
    "repo": "langfuse/langfuse",
    "author": "langfuse",
    "persona": "AI Ops & Tooling",
    "targetRole": "LLMOps & Observability Engineers",
    "stars": 35464,
    "wGrowth": 298.5,
    "mGrowth": 33.8,
    "website": "https://langfuse.com",
    "traffic": "580k/mo",
    "trafficNum": 580000,
    "searchVol": "38,000/mo",
    "searchVolNum": 38000,
    "searchTrend": "+78%",
    "latency": 32,
    "tokens": 95,
    "desc": "Open source LLM engineering platform: observability, metrics, evaluations, prompt management.",
    "useCase": "Inspect detailed step-by-step traces of multi-agent runs, latency bottlenecks, and token spend.",
    "arch": "Next.js, ClickHouse, PostgreSQL, OpenTelemetry.",
    "tags": [
      "observability",
      "tracing",
      "evals"
    ],
    "rank": 218,
    "elo": 1179,
    "eloCi": "±19",
    "winRate": 70,
    "starsDisplay": "35.5k",
    "weeklyStars": 27258,
    "weeklyGrowth": "+298.5%",
    "monthlyGrowth": "+33.8%",
    "wGrowthNum": 7.8,
    "mGrowthNum": 33.8,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/langfuse.png",
    "githubUrl": "https://github.com/langfuse/langfuse"
  },
  {
    "id": "cloudflare-d1",
    "name": "Cloudflare D1 SQL",
    "repo": "cloudflare/d1",
    "author": "cloudflare",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Full-Stack Edge Founders",
    "stars": 15600,
    "wGrowth": "+7.8%",
    "mGrowth": "+29.4%",
    "website": "https://developers.cloudflare.com/d1",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "65,000/mo",
    "searchVolNum": 65000,
    "searchTrend": "+51% (Strong)",
    "latency": "12ms",
    "tokens": "-30%",
    "desc": "Serverless edge relational SQLite database with automatic read replication across 300+ data centers for instantaneous response times.",
    "useCase": "Storing structured user data, agent memories, and session state with zero database provisioning and microsecond local queries.",
    "arch": "SQLite engine, Cloudflare distributed consensus, Read replication.",
    "tags": [
      "SQLite",
      "Edge Database",
      "Serverless SQL",
      "D1",
      "Cloudflare"
    ],
    "wGrowthNum": 7.8,
    "mGrowthNum": 29.4,
    "latencyNum": 12,
    "tokensNum": 30,
    "weeklyStars": 1216,
    "weeklyGrowth": "+7.8%",
    "monthlyGrowth": "+29.4%",
    "latencyMs": "12ms",
    "tokensSaved": "-30%",
    "starsDisplay": "15.6k",
    "logoUrl": "https://github.com/cloudflare.png",
    "githubUrl": "https://github.com/cloudflare/d1",
    "rank": 219,
    "elo": 1193,
    "eloCi": "±20",
    "winRate": 72
  },
  {
    "id": "hasura",
    "name": "Hasura GraphQL",
    "repo": "hasura/graphql-engine",
    "author": "hasura",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Full-Stack Founders",
    "stars": 32129,
    "wGrowth": 7.8,
    "mGrowth": 29.6,
    "website": "https://github.com/hasura/graphql-engine",
    "traffic": "1,575,000/mo",
    "trafficNum": 1575000,
    "searchVol": "315,000/mo",
    "searchVolNum": 315000,
    "searchTrend": "+140%",
    "latency": 25,
    "tokens": 90,
    "desc": "High-performance verified open-source agentic tool Hasura GraphQL powering production AI workloads.",
    "useCase": "Integrate Hasura GraphQL into agentic loops for verified deterministic execution and token efficiency.",
    "arch": "Modern open-source stack, MIT/Apache 2.0 license.",
    "tags": [
      "open-source",
      "agentic",
      "tool",
      "benchmark"
    ],
    "elo": 1311,
    "eloCi": "±20",
    "winRate": 85,
    "starsDisplay": "32.1k",
    "weeklyStars": 3086,
    "weeklyGrowth": "+7.8%",
    "monthlyGrowth": "+29.6%",
    "wGrowthNum": 7.8,
    "mGrowthNum": 29.6,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/hasura.png",
    "githubUrl": "https://github.com/hasura/graphql-engine",
    "rank": 220
  },
  {
    "id": "google-genai-sdk",
    "name": "Google Gen AI SDK",
    "repo": "google-gemini/generative-ai-python",
    "author": "google-gemini",
    "persona": "Researchers & STEM",
    "targetRole": "Gemini 2.5 Developers & Researchers",
    "stars": 2325,
    "wGrowth": "+7.5%",
    "mGrowth": "+28.1%",
    "website": "https://ai.google.dev",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+58% (Strong)",
    "latency": "40ms",
    "tokens": "-48%",
    "desc": "Official developer SDK for Gemini 2.5 Pro and Flash Thinking, supporting deep reasoning mode, Google Search grounding, and live code execution.",
    "useCase": "Building agentic workflows with native thinking traces, multi-modal audio/vision streaming, and zero-cost AI Studio integration.",
    "arch": "Python, Asyncio, gRPC, Google AI Studio / Vertex AI dual endpoints.",
    "tags": [
      "Gemini 2.5",
      "Thinking Mode",
      "Google AI",
      "Grounding",
      "Multi-Modal"
    ],
    "wGrowthNum": 7.5,
    "mGrowthNum": 28.1,
    "latencyNum": 40,
    "tokensNum": 48,
    "weeklyStars": 1110,
    "weeklyGrowth": "+7.5%",
    "monthlyGrowth": "+28.1%",
    "latencyMs": "40ms",
    "tokensSaved": "-48%",
    "starsDisplay": "2.3k",
    "logoUrl": "https://github.com/google-gemini.png",
    "githubUrl": "https://github.com/google-gemini/generative-ai-python",
    "rank": 221,
    "elo": 1189,
    "eloCi": "±21",
    "winRate": 72
  },
  {
    "id": "vllm",
    "name": "vLLM High-Throughput Serving",
    "repo": "vllm-project/vllm",
    "author": "vllm-project",
    "persona": "AI Ops & Tooling",
    "targetRole": "Inference & Platform Architects",
    "stars": 93309,
    "wGrowth": 162.1,
    "mGrowth": 31,
    "website": "https://vllm.ai",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+115%",
    "latency": 14,
    "tokens": 98,
    "desc": "A high-throughput and memory-efficient LLM serving engine powered by PagedAttention and continuous batching.",
    "useCase": "Serve self-hosted DeepSeek and Mistral endpoints with 24x higher request concurrency on private GPU fleets.",
    "arch": "Python, C++, CUDA kernels, PagedAttention, Ray cluster scaling.",
    "tags": [
      "high-throughput",
      "paged-attention",
      "llm-serving",
      "gpu-cluster"
    ],
    "rank": 222,
    "elo": 1228,
    "eloCi": "±22",
    "winRate": 77,
    "starsDisplay": "93.3k",
    "weeklyStars": 60272,
    "weeklyGrowth": "+162.1%",
    "monthlyGrowth": "+31.0%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 31,
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/vllm-project.png",
    "githubUrl": "https://github.com/vllm-project/vllm"
  },
  {
    "id": "aider-chat",
    "name": "Aider",
    "repo": "paul-gauthier/aider",
    "author": "paul-gauthier",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Senior Backend & Full-Stack SWEs",
    "stars": 49404,
    "wGrowth": 87.1,
    "mGrowth": 31,
    "website": "https://aider.chat",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+68%",
    "latency": 55,
    "tokens": 92,
    "desc": "Terminal-based AI pair programming tool with repo-wide tree-sitter map, automatic Git commits, and architect-editor modes.",
    "useCase": "Refactor multi-file architectures in terminal while Aider automatically stages and commits clean conventional commits per feature.",
    "arch": "Tree-sitter AST, GitPython, LiteLLM, Prompt Caching.",
    "tags": [
      "pair-programming",
      "git-commits",
      "terminal",
      "swe-bench"
    ],
    "rank": 223,
    "elo": 1209,
    "eloCi": "±23",
    "winRate": 74,
    "starsDisplay": "49.4k",
    "weeklyStars": 24904,
    "weeklyGrowth": "+87.1%",
    "monthlyGrowth": "+31.0%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 31,
    "latencyMs": "55ms",
    "latencyNum": 55,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/paul-gauthier.png",
    "githubUrl": "https://github.com/paul-gauthier/aider"
  },
  {
    "id": "uvr5",
    "name": "Ultimate Vocal Remover",
    "repo": "Anjok07/ultimatevocalremovergui",
    "author": "Anjok07",
    "persona": "Music & Audio Creators",
    "targetRole": "DJs & Audio Engineers",
    "stars": 26501,
    "wGrowth": 33.8,
    "mGrowth": 31.5,
    "website": "https://ultimatevocalremover.com",
    "traffic": "950k/mo",
    "trafficNum": 950000,
    "searchVol": "180,000/mo",
    "searchVolNum": 180000,
    "searchTrend": "+52%",
    "latency": 160,
    "tokens": 86,
    "desc": "The gold standard core desktop and CLI tool for acoustic ensemble separation using MDX-Net, VR Architecture, and Demucs v4.",
    "useCase": "Extract studio-grade acapellas with zero reverb bleed and remove crowd noise from live concert recordings.",
    "arch": "MDX-Net, ONNX, PyTorch, Soundfile, Tkinter.",
    "tags": [
      "vocal-removal",
      "acapella",
      "dj-mix",
      "audio-cleaning"
    ],
    "rank": 224,
    "elo": 1196,
    "eloCi": "±24",
    "winRate": 73,
    "starsDisplay": "26.5k",
    "weeklyStars": 8126,
    "weeklyGrowth": "+33.8%",
    "monthlyGrowth": "+31.5%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 31.5,
    "latencyMs": "160ms",
    "latencyNum": 160,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "logoUrl": "https://github.com/Anjok07.png",
    "githubUrl": "https://github.com/Anjok07/ultimatevocalremovergui"
  },
  {
    "id": "documenso",
    "name": "Documenso",
    "repo": "documenso/documenso",
    "author": "documenso",
    "persona": "Founders & Indie Hackers",
    "targetRole": "B2B Founders & Legal",
    "stars": 15349,
    "wGrowth": 56.6,
    "mGrowth": 31,
    "website": "https://documenso.com",
    "traffic": "320k/mo",
    "trafficNum": 320000,
    "searchVol": "24,000/mo",
    "searchVolNum": 24000,
    "searchTrend": "+62%",
    "latency": 48,
    "tokens": 88,
    "desc": "Open-source DocuSign alternative for creating, signing, and managing legally-binding electronic document workflows.",
    "useCase": "Automate founder partnership contracts, enterprise pilots, and investor SAFE agreements with auditable cryptographic certificates.",
    "arch": "Next.js, Prisma, PostgreSQL, PDF-Lib, Tailwind CSS.",
    "tags": [
      "signatures",
      "contracts",
      "docusign-alt",
      "b2b-deals"
    ],
    "rank": 225,
    "elo": 1176,
    "eloCi": "±25",
    "winRate": 70,
    "starsDisplay": "15.3k",
    "weeklyStars": 6254,
    "weeklyGrowth": "+56.6%",
    "monthlyGrowth": "+31.0%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 31,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/documenso.png",
    "githubUrl": "https://github.com/documenso/documenso"
  },
  {
    "id": "e2b",
    "name": "E2B Code Sandbox",
    "repo": "e2b-dev/E2B",
    "author": "e2b-dev",
    "persona": "Autonomous Software Engineers",
    "targetRole": "AI Sandbox Engineers",
    "stars": 14209,
    "wGrowth": 92,
    "mGrowth": 31,
    "website": "https://e2b.dev",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+74%",
    "latency": 45,
    "tokens": 93,
    "desc": "Isolated cloud sandboxes for running untrusted AI-generated code safely.",
    "useCase": "Execute agent-written Python charts, terminal bash scripts, and builds in micro-VMs.",
    "arch": "Firecracker microVM, Linux kernel, REST API.",
    "tags": [
      "sandbox",
      "code-execution",
      "firecracker"
    ],
    "rank": 226,
    "elo": 1171,
    "eloCi": "±26",
    "winRate": 69,
    "starsDisplay": "14.2k",
    "weeklyStars": 7341,
    "weeklyGrowth": "+92%",
    "monthlyGrowth": "+31.0%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 31,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/e2b-dev.png",
    "githubUrl": "https://github.com/e2b-dev/E2B"
  },
  {
    "id": "prisma",
    "name": "Prisma ORM",
    "repo": "prisma/prisma",
    "author": "prisma",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Full-Stack TypeScript Founders",
    "stars": 47699,
    "wGrowth": 15.8,
    "mGrowth": 28,
    "website": "https://prisma.io",
    "traffic": "3.8M/mo",
    "trafficNum": 3800000,
    "searchVol": "480,000/mo",
    "searchVolNum": 480000,
    "searchTrend": "+95%",
    "latency": 25,
    "tokens": 88,
    "desc": "Next-generation ORM for Node.js and TypeScript. Declarative schema, automated migrations, and fully typesafe client.",
    "useCase": "Deliver complete database access layers with compile-time type safety across PostgreSQL, MySQL, and SQLite.",
    "arch": "Rust query engine, TypeScript client, Schema AST parser.",
    "tags": [
      "orm",
      "typescript",
      "database",
      "fullstack"
    ],
    "elo": 1300,
    "eloCi": "±20",
    "winRate": 85,
    "logoUrl": "https://github.com/prisma.png",
    "githubUrl": "https://github.com/prisma/prisma",
    "starsDisplay": "47.7k",
    "weeklyStars": 9465,
    "weeklyGrowth": "+15.8%",
    "monthlyGrowth": "+28.0%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 28,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 227
  },
  {
    "id": "langchain",
    "name": "LangChain",
    "repo": "langchain-ai/langchain",
    "author": "langchain-ai",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Enterprise Agent Architects",
    "stars": 147522,
    "wGrowth": 51.3,
    "mGrowth": 28,
    "website": "https://langchain.com",
    "traffic": "4.8M/mo",
    "searchVol": "680,000/mo",
    "latency": 32,
    "tokens": 82,
    "desc": "Framework for developing applications powered by large language models, agents, and tool calling.",
    "useCase": "Assemble multi-model pipelines, memory components, and vector database connectors.",
    "arch": "Python/TypeScript, LCEL expression language, Pydantic.",
    "tags": [
      "agent-framework",
      "llm-orchestration",
      "lcel",
      "enterprise-ai"
    ],
    "starsDisplay": "147.5k",
    "weeklyStars": 57042,
    "weeklyGrowth": "+51.3%",
    "monthlyGrowth": "+28%",
    "wGrowthNum": 7.2,
    "mGrowthNum": 28,
    "trafficNum": 4000000,
    "searchVolNum": 680000,
    "searchTrend": "+72%",
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-82%",
    "tokensNum": 82,
    "elo": 1280,
    "eloCi": "±18",
    "winRate": 82,
    "logoUrl": "https://github.com/langchain-ai.png",
    "githubUrl": "https://github.com/langchain-ai/langchain",
    "rank": 228
  },
  {
    "id": "litellm",
    "name": "LiteLLM",
    "repo": "BerriAI/litellm",
    "author": "BerriAI",
    "persona": "AI Ops & Tooling",
    "targetRole": "Backend & Cost Engineers",
    "stars": 60262,
    "wGrowth": 227.5,
    "mGrowth": 30.5,
    "website": "https://litellm.ai",
    "traffic": "920k/mo",
    "trafficNum": 920000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+68%",
    "latency": 28,
    "tokens": 96,
    "desc": "Call 100+ LLMs using the OpenAI input/output format. Centralized spend tracking, load balancing, rate limiting, and circuit breakers.",
    "useCase": "Prevent downtime by automatically failing over from Anthropic Claude to Gemini 2.5 Pro or DeepSeek if rate limits or 500 errors hit.",
    "arch": "Python async proxy, Redis cache, OpenTelemetry instrumentation.",
    "tags": [
      "proxy",
      "failover",
      "cost-tracking",
      "multi-provider"
    ],
    "rank": 229,
    "elo": 1192,
    "eloCi": "±18",
    "winRate": 72,
    "starsDisplay": "60.3k",
    "weeklyStars": 43168,
    "weeklyGrowth": "+227.5%",
    "monthlyGrowth": "+30.5%",
    "wGrowthNum": 7.1,
    "mGrowthNum": 30.5,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/BerriAI.png",
    "githubUrl": "https://github.com/BerriAI/litellm"
  },
  {
    "id": "formbricks",
    "name": "Formbricks",
    "repo": "formbricks/formbricks",
    "author": "formbricks",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Product Managers & Founders",
    "stars": 13066,
    "wGrowth": 46.8,
    "mGrowth": 30.5,
    "website": "https://formbricks.com",
    "traffic": "380k/mo",
    "trafficNum": 380000,
    "searchVol": "18,000/mo",
    "searchVolNum": 18000,
    "searchTrend": "+64%",
    "latency": 46,
    "tokens": 88,
    "desc": "Open-source survey suite for targeted user feedback and customer satisfaction tracking.",
    "useCase": "Run micro-surveys at point of churn or post-upgrade to collect qualitative feedback.",
    "arch": "Next.js, Prisma, PostgreSQL, Tailwind.",
    "tags": [
      "surveys",
      "feedback",
      "churn",
      "cro"
    ],
    "rank": 230,
    "elo": 1173,
    "eloCi": "±19",
    "winRate": 69,
    "starsDisplay": "13.1k",
    "weeklyStars": 4797,
    "weeklyGrowth": "+46.8%",
    "monthlyGrowth": "+30.5%",
    "wGrowthNum": 7.1,
    "mGrowthNum": 30.5,
    "latencyMs": "46ms",
    "latencyNum": 46,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/formbricks.png",
    "githubUrl": "https://github.com/formbricks/formbricks"
  },
  {
    "id": "payload-cms",
    "name": "Payload CMS",
    "repo": "payloadcms/payload",
    "author": "payloadcms",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Content & SaaS Founders",
    "stars": 45120,
    "wGrowth": 53.5,
    "mGrowth": 30.5,
    "website": "https://payloadcms.com",
    "traffic": "780k/mo",
    "trafficNum": 780000,
    "searchVol": "65,000/mo",
    "searchVolNum": 65000,
    "searchTrend": "+68%",
    "latency": 36,
    "tokens": 92,
    "desc": "The open-source full-stack Next.js database and headless CMS. 100% TypeScript with native draft preview and SEO metadata generation.",
    "useCase": "Manage programmatically generated landing pages, blogs, and customer briefs directly within your main Next.js App Router repository.",
    "arch": "Next.js App Router, Postgres/MongoDB, Lexical Rich Text, Tailwind.",
    "tags": [
      "nextjs",
      "cms",
      "seo-pages",
      "typescript"
    ],
    "rank": 231,
    "elo": 1213,
    "eloCi": "±20",
    "winRate": 75,
    "starsDisplay": "45.1k",
    "weeklyStars": 17748,
    "weeklyGrowth": "+53.5%",
    "monthlyGrowth": "+30.5%",
    "wGrowthNum": 6.9,
    "mGrowthNum": 30.5,
    "latencyMs": "36ms",
    "latencyNum": 36,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/payloadcms.png",
    "githubUrl": "https://github.com/payloadcms/payload"
  },
  {
    "id": "whisperx",
    "name": "WhisperX",
    "repo": "m-bain/whisperX",
    "author": "m-bain",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Subtitlers & Musicians",
    "stars": 24401,
    "wGrowth": 71.8,
    "mGrowth": 29.5,
    "website": "https://github.com/m-bain/whisperX",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "68,000/mo",
    "searchVolNum": 68000,
    "searchTrend": "+58%",
    "latency": 50,
    "tokens": 94,
    "desc": "Fast automatic speech recognition with word-level forced alignment and diarization using Whisper and wav2vec2.",
    "useCase": "Synchronize lyric captions precisely to singing vocals for music videos.",
    "arch": "PyTorch, wav2vec2, Faster-Whisper, PyAnnote.",
    "tags": [
      "forced-alignment",
      "lyrics-sync",
      "diarization"
    ],
    "rank": 232,
    "elo": 1182,
    "eloCi": "±22",
    "winRate": 71,
    "starsDisplay": "24.4k",
    "weeklyStars": 11166,
    "weeklyGrowth": "+71.8%",
    "monthlyGrowth": "+29.5%",
    "wGrowthNum": 6.8,
    "mGrowthNum": 29.5,
    "latencyMs": "50ms",
    "latencyNum": 50,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/m-bain.png",
    "githubUrl": "https://github.com/m-bain/whisperX"
  },
  {
    "id": "instructor",
    "name": "Instructor",
    "repo": "jxnl/instructor",
    "author": "jxnl",
    "persona": "Researchers & STEM",
    "targetRole": "Research Data Scientists",
    "stars": 13981,
    "wGrowth": 70.5,
    "mGrowth": 29.5,
    "website": "https://python.useinstructor.com",
    "traffic": "410k/mo",
    "trafficNum": 410000,
    "searchVol": "36,000/mo",
    "searchVolNum": 36000,
    "searchTrend": "+64%",
    "latency": 38,
    "tokens": 96,
    "desc": "Structured outputs with Pydantic for Python and TypeScript. Guarantees 100% typed, validated JSON from any language model.",
    "useCase": "Extract dense scientific papers into structured DAG ontology models with automatic validation retries on schema errors.",
    "arch": "Pydantic v2, Functional calling, Tenacity retries, Zod.",
    "tags": [
      "structured-data",
      "pydantic",
      "typed-output",
      "validation"
    ],
    "rank": 233,
    "elo": 1170,
    "eloCi": "±23",
    "winRate": 69,
    "starsDisplay": "14.0k",
    "weeklyStars": 6338,
    "weeklyGrowth": "+70.5%",
    "monthlyGrowth": "+29.5%",
    "wGrowthNum": 6.8,
    "mGrowthNum": 29.5,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/jxnl.png",
    "githubUrl": "https://github.com/jxnl/instructor"
  },
  {
    "id": "promptfoo",
    "name": "Promptfoo",
    "repo": "promptfoo/promptfoo",
    "author": "promptfoo",
    "persona": "Researchers & STEM",
    "targetRole": "AI Safety & Eval Leads",
    "stars": 25773,
    "wGrowth": 315.7,
    "mGrowth": 29.5,
    "website": "https://promptfoo.dev",
    "traffic": "380k/mo",
    "trafficNum": 380000,
    "searchVol": "28,000/mo",
    "searchVolNum": 28000,
    "searchTrend": "+68%",
    "latency": 38,
    "tokens": 95,
    "desc": "Test and benchmark LLM outputs, red-team prompts, and catch hallucinations.",
    "useCase": "Automated assertion testing of agent responses across adversarial prompts in CI/CD.",
    "arch": "TypeScript, Node.js, CLI runner.",
    "tags": [
      "evaluation",
      "red-teaming",
      "llm-test"
    ],
    "rank": 234,
    "elo": 1166,
    "eloCi": "±24",
    "winRate": 68,
    "starsDisplay": "25.8k",
    "weeklyStars": 19994,
    "weeklyGrowth": "+315.7%",
    "monthlyGrowth": "+29.5%",
    "wGrowthNum": 6.8,
    "mGrowthNum": 29.5,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/promptfoo.png",
    "githubUrl": "https://github.com/promptfoo/promptfoo"
  },
  {
    "id": "openai-swarm",
    "name": "OpenAI Swarm",
    "repo": "openai/swarm",
    "author": "openai",
    "persona": "Researchers & STEM",
    "targetRole": "Multi-Agent Researchers",
    "stars": 22039,
    "wGrowth": null,
    "mGrowth": "+24.5%",
    "website": "https://github.com/openai/swarm",
    "traffic": "650k/mo",
    "trafficNum": 650000,
    "searchVol": "72,000/mo",
    "searchVolNum": 72000,
    "searchTrend": "+35% (Steady)",
    "latency": "55ms",
    "tokens": "-40%",
    "desc": "Ergonomic educational framework for multi-agent orchestration, agent handoffs, and deterministic routine composition.",
    "useCase": "Studying and prototyping lightweight multi-agent delegation patterns with transparent Python code and minimal abstractions.",
    "arch": "Python, Stateless agent functions, OpenAI Chat Completions tool calling.",
    "tags": [
      "Multi-Agent",
      "Agent Handoffs",
      "OpenAI",
      "Routines",
      "Coordination"
    ],
    "wGrowthNum": 6.8,
    "mGrowthNum": 24.5,
    "latencyNum": 55,
    "tokensNum": 40,
    "weeklyStars": 6008,
    "weeklyGrowth": "+NaN%",
    "monthlyGrowth": "+24.5%",
    "latencyMs": "55ms",
    "tokensSaved": "-40%",
    "starsDisplay": "22.0k",
    "logoUrl": "https://github.com/openai.png",
    "githubUrl": "https://github.com/openai/swarm",
    "rank": 235,
    "elo": 1188,
    "eloCi": "±25",
    "winRate": 71
  },
  {
    "id": "puppeteer",
    "name": "Puppeteer",
    "repo": "puppeteer/puppeteer",
    "author": "puppeteer",
    "persona": "AI Ops & Tooling",
    "targetRole": "Scraping & Automation Engineers",
    "stars": 95667,
    "wGrowth": 7.2,
    "mGrowth": 26,
    "website": "https://pptr.dev",
    "traffic": "4.2M/mo",
    "trafficNum": 4200000,
    "searchVol": "590,000/mo",
    "searchVolNum": 590000,
    "searchTrend": "+90%",
    "latency": 28,
    "tokens": 88,
    "desc": "Node.js library which provides a high-level API to control Chrome or Firefox over the DevTools Protocol.",
    "useCase": "Drive automated browser workflows, PDF generation, and automated screenshots in serverless environments.",
    "arch": "TypeScript, Chrome DevTools Protocol (CDP), WebSocket transport.",
    "tags": [
      "browser-automation",
      "cdp",
      "web-scraping",
      "headless-chrome"
    ],
    "elo": 1320,
    "eloCi": "±18",
    "winRate": 87,
    "logoUrl": "https://github.com/puppeteer.png",
    "githubUrl": "https://github.com/puppeteer/puppeteer",
    "starsDisplay": "95.7k",
    "weeklyStars": 12533,
    "weeklyGrowth": "+7.2%",
    "monthlyGrowth": "+26.0%",
    "wGrowthNum": 6.8,
    "mGrowthNum": 26,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "rank": 236
  },
  {
    "id": "dub-sh",
    "name": "Dub.co Link Hub",
    "repo": "dubinc/dub",
    "author": "dubinc",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Growth Founders",
    "stars": 24867,
    "wGrowth": 23.1,
    "mGrowth": 28,
    "website": "https://dub.co",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "48,000/mo",
    "searchVolNum": 48000,
    "searchTrend": "+55%",
    "latency": 35,
    "tokens": 93,
    "desc": "Open-source link management and attribution engine with geo-targeting, conversion tracking, custom domain routing, and analytics.",
    "useCase": "Track referral conversions, create dynamic social share cards, and route affiliate campaigns to localized checkout flows.",
    "arch": "Next.js App Router, Tinybird ClickHouse analytics, Upstash Redis, Prisma, Tailwind CSS.",
    "tags": [
      "shortlinks",
      "attribution",
      "conversion",
      "marketing-funnel"
    ],
    "rank": 237,
    "elo": 1192,
    "eloCi": "±26",
    "winRate": 72,
    "starsDisplay": "24.9k",
    "weeklyStars": 5980,
    "weeklyGrowth": "+23.1%",
    "monthlyGrowth": "+28.0%",
    "wGrowthNum": 6.5,
    "mGrowthNum": 28,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/dubinc.png",
    "githubUrl": "https://github.com/dubinc/dub"
  },
  {
    "id": "rvc-webui",
    "name": "RVC Voice Conversion",
    "repo": "RVC-Project/Retrieval-based-Voice-Conversion-WebUI",
    "author": "RVC-Project",
    "persona": "Music & Audio Creators",
    "targetRole": "Vocalists & Audio Experimenters",
    "stars": 38624,
    "wGrowth": 44.1,
    "mGrowth": 27.5,
    "website": "https://github.com/RVC-Project",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "210,000/mo",
    "searchVolNum": 210000,
    "searchTrend": "+48%",
    "latency": 140,
    "tokens": 87,
    "desc": "Retrieval-based voice conversion tool that transforms vocal singing tracks into different target character voices with zero pitch glitch.",
    "useCase": "Convert a reference singing track into a smooth soprano or synthetic vocal tone with pitch preservation.",
    "arch": "Hubert feature extraction, FAISS vector retrieval, VITS acoustic model.",
    "tags": [
      "voice-conversion",
      "singing-voice",
      "rvc",
      "vocal-synth"
    ],
    "rank": 238,
    "elo": 1204,
    "eloCi": "±18",
    "winRate": 74,
    "starsDisplay": "38.6k",
    "weeklyStars": 13539,
    "weeklyGrowth": "+44.1%",
    "monthlyGrowth": "+27.5%",
    "wGrowthNum": 6.4,
    "mGrowthNum": 27.5,
    "latencyMs": "140ms",
    "latencyNum": 140,
    "tokensSaved": "-87%",
    "tokensNum": 87,
    "logoUrl": "https://github.com/RVC-Project.png",
    "githubUrl": "https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI"
  },
  {
    "id": "openstatus",
    "name": "OpenStatus",
    "repo": "openstatusHQ/openstatus",
    "author": "openstatusHQ",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Reliability & Founders",
    "stars": 9172,
    "wGrowth": 69.9,
    "mGrowth": 27.5,
    "website": "https://openstatus.dev",
    "traffic": "210k/mo",
    "trafficNum": 210000,
    "searchVol": "12,000/mo",
    "searchVolNum": 12000,
    "searchTrend": "+52%",
    "latency": 36,
    "tokens": 92,
    "desc": "Open source synthetic monitoring and beautiful status pages for modern SaaS.",
    "useCase": "Display verified uptime badges on landing pages to build trust with enterprise customers.",
    "arch": "Next.js, Tinybird, Cloudflare Workers, Tailwind.",
    "tags": [
      "uptime",
      "monitoring",
      "status-page"
    ],
    "rank": 239,
    "elo": 1161,
    "eloCi": "±19",
    "winRate": 68,
    "starsDisplay": "9.2k",
    "weeklyStars": 4117,
    "weeklyGrowth": "+69.9%",
    "monthlyGrowth": "+27.5%",
    "wGrowthNum": 6.4,
    "mGrowthNum": 27.5,
    "latencyMs": "36ms",
    "latencyNum": 36,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/openstatusHQ.png",
    "githubUrl": "https://github.com/openstatusHQ/openstatus"
  },
  {
    "id": "openhands",
    "name": "OpenHands",
    "repo": "All-Hands-AI/OpenHands",
    "author": "All-Hands-AI",
    "persona": "Autonomous Software Engineers",
    "targetRole": "DevOps & Core SWEs",
    "stars": 90143,
    "wGrowth": 118.8,
    "mGrowth": 27,
    "website": "https://all-hands.dev",
    "traffic": "980k/mo",
    "trafficNum": 980000,
    "searchVol": "115,000/mo",
    "searchVolNum": 115000,
    "searchTrend": "+55%",
    "latency": 75,
    "tokens": 86,
    "desc": "Open-source platform for autonomous software development agents in isolated Docker sandboxes, scoring state-of-the-art on SWE-bench.",
    "useCase": "Resolve real GitHub issues by cloning the repo in Docker, reproducing the bug with a unit test, and proposing a verified PR.",
    "arch": "Docker Sandbox, Event Stream, FastAPI, Playwright, React.",
    "tags": [
      "docker-sandbox",
      "swe-bench",
      "bugfix-agent",
      "autonomous"
    ],
    "rank": 240,
    "elo": 1231,
    "eloCi": "±20",
    "winRate": 78,
    "starsDisplay": "90.1k",
    "weeklyStars": 51497,
    "weeklyGrowth": "+118.8%",
    "monthlyGrowth": "+27.0%",
    "wGrowthNum": 6.2,
    "mGrowthNum": 27,
    "latencyMs": "75ms",
    "latencyNum": 75,
    "tokensSaved": "-86%",
    "tokensNum": 86,
    "logoUrl": "https://github.com/All-Hands-AI.png",
    "githubUrl": "https://github.com/All-Hands-AI/OpenHands"
  },
  {
    "id": "ast-grep",
    "name": "ast-grep",
    "repo": "ast-grep/ast-grep",
    "author": "ast-grep",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Refactoring Specialists",
    "stars": 16131,
    "wGrowth": 81.2,
    "mGrowth": 26.5,
    "website": "https://ast-grep.github.io",
    "traffic": "340k/mo",
    "trafficNum": 340000,
    "searchVol": "26,000/mo",
    "searchVolNum": 26000,
    "searchTrend": "+52%",
    "latency": 18,
    "tokens": 98,
    "desc": "A fast and polyglot tool for code searching, linting, and rewriting using ASTs.",
    "useCase": "Safely execute codebase refactors matching syntax patterns instead of regex.",
    "arch": "Rust, Tree-sitter, AST structural matcher.",
    "tags": [
      "ast",
      "tree-sitter",
      "refactor",
      "static-analysis"
    ],
    "rank": 241,
    "elo": 1166,
    "eloCi": "±21",
    "winRate": 68,
    "starsDisplay": "16.1k",
    "weeklyStars": 7782,
    "weeklyGrowth": "+81.2%",
    "monthlyGrowth": "+26.5%",
    "wGrowthNum": 6.2,
    "mGrowthNum": 26.5,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/ast-grep.png",
    "githubUrl": "https://github.com/ast-grep/ast-grep"
  },
  {
    "id": "growthbook",
    "name": "GrowthBook",
    "repo": "growthbook/growthbook",
    "author": "growthbook",
    "persona": "Marketers & Growth",
    "targetRole": "Conversion Rate Specialists",
    "stars": 8480,
    "wGrowth": 32.5,
    "mGrowth": 26.5,
    "website": "https://growthbook.io",
    "traffic": "350k/mo",
    "trafficNum": 350000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+52%",
    "latency": 35,
    "tokens": 93,
    "desc": "Open-source A/B testing and feature flagging platform with Bayesian statistics engine and warehouse-native queries.",
    "useCase": "Run headline and CTA split tests directly against your Snowflake or BigQuery data without sending user PII to third parties.",
    "arch": "Next.js, Python statistical engine, Mongo/Postgres.",
    "tags": [
      "conversion-rate",
      "ab-testing",
      "feature-flags",
      "cro"
    ],
    "rank": 242,
    "elo": 1161,
    "eloCi": "±22",
    "winRate": 68,
    "starsDisplay": "8.5k",
    "weeklyStars": 2476,
    "weeklyGrowth": "+32.5%",
    "monthlyGrowth": "+26.5%",
    "wGrowthNum": 6.2,
    "mGrowthNum": 26.5,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/growthbook.png",
    "githubUrl": "https://github.com/growthbook/growthbook"
  },
  {
    "id": "outlines",
    "name": "Outlines",
    "repo": "dottxt-ai/outlines",
    "author": "dottxt-ai",
    "persona": "Researchers & STEM",
    "targetRole": "Theoretical Computer Scientists",
    "stars": 15905,
    "wGrowth": 28.3,
    "mGrowth": 26.5,
    "website": "https://dottxt.ai",
    "traffic": "320k/mo",
    "trafficNum": 320000,
    "searchVol": "44,000/mo",
    "searchVolNum": 44000,
    "searchTrend": "+55%",
    "latency": 35,
    "tokens": 97,
    "desc": "Fast, guaranteed structured generation with Context-Free Grammars and Regular Expressions directly masking model token logits.",
    "useCase": "Ensure an LLM generates syntactically valid SQL queries or Lean 4 proof commands without ever producing a syntax error.",
    "arch": "Rust regex engine, FSM logit masking, PyTorch, vLLM integration.",
    "tags": [
      "cfg-grammars",
      "logit-masking",
      "deterministic",
      "regex"
    ],
    "rank": 243,
    "elo": 1172,
    "eloCi": "±24",
    "winRate": 69,
    "starsDisplay": "15.9k",
    "weeklyStars": 4261,
    "weeklyGrowth": "+28.3%",
    "monthlyGrowth": "+26.5%",
    "wGrowthNum": 6.1,
    "mGrowthNum": 26.5,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/dottxt-ai.png",
    "githubUrl": "https://github.com/dottxt-ai/outlines"
  },
  {
    "id": "inngest",
    "name": "Inngest Workflows",
    "repo": "inngest/inngest",
    "author": "inngest",
    "persona": "Marketers & Growth",
    "targetRole": "Growth Engineers",
    "stars": 5923,
    "wGrowth": 41,
    "mGrowth": 26.2,
    "website": "https://inngest.com",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "18,000/mo",
    "searchVolNum": 18000,
    "searchTrend": "+55%",
    "latency": 38,
    "tokens": 92,
    "desc": "Event-driven durable execution engine for background jobs, email workflows, and webhooks.",
    "useCase": "Orchestrate 7-day automated drip campaigns that pause when users activate a key feature.",
    "arch": "Go server, TypeScript SDK, Event streams.",
    "tags": [
      "event-driven",
      "drip-campaigns",
      "lifecycle"
    ],
    "rank": 244,
    "elo": 1156,
    "eloCi": "±25",
    "winRate": 67,
    "starsDisplay": "5.9k",
    "weeklyStars": 1979,
    "weeklyGrowth": "+41%",
    "monthlyGrowth": "+26.2%",
    "wGrowthNum": 6.1,
    "mGrowthNum": 26.2,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/inngest.png",
    "githubUrl": "https://github.com/inngest/inngest"
  },
  {
    "id": "dspy",
    "name": "DSPy",
    "repo": "stanfordnlp/dspy",
    "author": "stanfordnlp",
    "persona": "Researchers & STEM",
    "targetRole": "NLP & AI Researchers",
    "stars": 38533,
    "wGrowth": 72,
    "mGrowth": 25.5,
    "website": "https://dspy.ai",
    "traffic": "680k/mo",
    "trafficNum": 680000,
    "searchVol": "72,000/mo",
    "searchVolNum": 72000,
    "searchTrend": "+52%",
    "latency": 62,
    "tokens": 94,
    "desc": "Stanford framework for algorithmically optimizing LM prompts and weights. Replaces fragile hand-written prompts with compiled modules.",
    "useCase": "Compile complex RAG pipelines where teleprompter optimizers automatically tune few-shot exemplars to maximize validation accuracy.",
    "arch": "MIPROv2 optimizer, BootstrapFewShot, Python declarative API.",
    "tags": [
      "prompt-compiler",
      "stanford",
      "teleprompter",
      "declarative-ai"
    ],
    "rank": 245,
    "elo": 1191,
    "eloCi": "±26",
    "winRate": 72,
    "starsDisplay": "38.5k",
    "weeklyStars": 17454,
    "weeklyGrowth": "+72%",
    "monthlyGrowth": "+25.5%",
    "wGrowthNum": 5.9,
    "mGrowthNum": 25.5,
    "latencyMs": "62ms",
    "latencyNum": 62,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/stanfordnlp.png",
    "githubUrl": "https://github.com/stanfordnlp/dspy"
  },
  {
    "id": "demucs",
    "name": "Demucs",
    "repo": "facebookresearch/demucs",
    "author": "facebookresearch",
    "persona": "Music & Audio Creators",
    "targetRole": "Music Producers & Audio Engineers",
    "stars": 10368,
    "wGrowth": 10.3,
    "mGrowth": 25,
    "website": "https://github.com/facebookresearch/demucs",
    "traffic": "680k/mo",
    "trafficNum": 680000,
    "searchVol": "140,000/mo",
    "searchVolNum": 140000,
    "searchTrend": "+42%",
    "latency": 180,
    "tokens": 88,
    "desc": "Deep learning music source separation by Meta AI. Separates any audio track into isolated vocals, drums, bass, and other instruments.",
    "useCase": "Isolate pristine vocal acapellas from master mixes to create instant remixes, karaoke backing tracks, or sample packs.",
    "arch": "PyTorch, Hybrid Transformer-Demucs, Torchaudio, CUDA.",
    "tags": [
      "stem-separation",
      "vocals-isolated",
      "meta-ai",
      "music-production"
    ],
    "rank": 246,
    "elo": 1164,
    "eloCi": "±18",
    "winRate": 68,
    "starsDisplay": "10.4k",
    "weeklyStars": 1513,
    "weeklyGrowth": "+10.3%",
    "monthlyGrowth": "+25.0%",
    "wGrowthNum": 5.8,
    "mGrowthNum": 25,
    "latencyMs": "180ms",
    "latencyNum": 180,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/facebookresearch.png",
    "githubUrl": "https://github.com/facebookresearch/demucs"
  },
  {
    "id": "ollama",
    "name": "Ollama Local Runner",
    "repo": "ollama/ollama",
    "author": "ollama",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Local AI & Backend Engineers",
    "stars": 182430,
    "wGrowth": 62.9,
    "mGrowth": 24,
    "website": "https://ollama.com",
    "traffic": "5.4M/mo",
    "trafficNum": 5400000,
    "searchVol": "550,000/mo",
    "searchVolNum": 550000,
    "searchTrend": "+95%",
    "latency": 22,
    "tokens": 97,
    "desc": "Get up and running with Llama 3, DeepSeek, Mistral, and Gemma models locally with a single command.",
    "useCase": "Run private, zero-cost code completion and automated test generation models directly on developer workstations without cloud egress.",
    "arch": "Go runtime, llama.cpp backend, Metal/CUDA acceleration, REST & Modelfile APIs.",
    "tags": [
      "local-llm",
      "offline-ai",
      "llama3",
      "developer-productivity"
    ],
    "rank": 247,
    "elo": 1365,
    "eloCi": "±19",
    "winRate": 92,
    "starsDisplay": "182.4k",
    "weeklyStars": 76701,
    "weeklyGrowth": "+62.9%",
    "monthlyGrowth": "+24.0%",
    "wGrowthNum": 5.6,
    "mGrowthNum": 24,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/ollama.png",
    "githubUrl": "https://github.com/ollama/ollama"
  },
  {
    "id": "infisical",
    "name": "Infisical",
    "repo": "Infisical/infisical",
    "author": "Infisical",
    "persona": "Founders & Indie Hackers",
    "targetRole": "DevOps & Founders",
    "stars": 29639,
    "wGrowth": 83,
    "mGrowth": 23.2,
    "website": "https://infisical.com",
    "traffic": "680k/mo",
    "trafficNum": 680000,
    "searchVol": "38,000/mo",
    "searchVolNum": 38000,
    "searchTrend": "+48%",
    "latency": 35,
    "tokens": 93,
    "desc": "Open source secret management platform. Centralize and sync env variables across teams.",
    "useCase": "Safely sync API keys across dev, staging, and production VPS environments without plaintext leaks.",
    "arch": "Node.js, Express, MongoDB, Vault encryption.",
    "tags": [
      "secrets",
      "env",
      "security",
      "devops"
    ],
    "rank": 248,
    "elo": 1175,
    "eloCi": "±20",
    "winRate": 70,
    "starsDisplay": "29.6k",
    "weeklyStars": 14313,
    "weeklyGrowth": "+83%",
    "monthlyGrowth": "+23.2%",
    "wGrowthNum": 5.4,
    "mGrowthNum": 23.2,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/Infisical.png",
    "githubUrl": "https://github.com/Infisical/infisical"
  },
  {
    "id": "continue-dev",
    "name": "Continue",
    "repo": "continuedev/continue",
    "author": "continuedev",
    "persona": "Autonomous Software Engineers",
    "targetRole": "IDE Power Users",
    "stars": 36144,
    "wGrowth": 65.8,
    "mGrowth": 22.5,
    "website": "https://continue.dev",
    "traffic": "950k/mo",
    "trafficNum": 950000,
    "searchVol": "85,000/mo",
    "searchVolNum": 85000,
    "searchTrend": "+38%",
    "latency": 45,
    "tokens": 93,
    "desc": "The leading open-source AI code assistant for VS Code and JetBrains.",
    "useCase": "Connect local Ollama models directly to your IDE for private autocomplete.",
    "arch": "TypeScript, VSCode Extension, Context Providers.",
    "tags": [
      "copilot-alt",
      "ide",
      "vscode",
      "autocomplete"
    ],
    "rank": 249,
    "elo": 1184,
    "eloCi": "±21",
    "winRate": 71,
    "starsDisplay": "36.1k",
    "weeklyStars": 15477,
    "weeklyGrowth": "+65.8%",
    "monthlyGrowth": "+22.5%",
    "wGrowthNum": 5.2,
    "mGrowthNum": 22.5,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/continuedev.png",
    "githubUrl": "https://github.com/continuedev/continue"
  },
  {
    "id": "ragas",
    "name": "Ragas",
    "repo": "explodinggradients/ragas",
    "author": "explodinggradients",
    "persona": "Researchers & STEM",
    "targetRole": "RAG Evaluation Engineers",
    "stars": 15950,
    "wGrowth": 69.7,
    "mGrowth": 22.4,
    "website": "https://ragas.io",
    "traffic": "520k/mo",
    "trafficNum": 520000,
    "searchVol": "45,000/mo",
    "searchVolNum": 45000,
    "searchTrend": "+48%",
    "latency": 45,
    "tokens": 93,
    "desc": "Supervised evaluation framework for Retrieval Augmented Generation (RAG) pipelines.",
    "useCase": "Quantify context recall, faithfulness, and answer relevance on production Q&A datasets.",
    "arch": "Python, Hugging Face Datasets, LangChain.",
    "tags": [
      "rag-eval",
      "metrics",
      "hallucination"
    ],
    "rank": 250,
    "elo": 1159,
    "eloCi": "±22",
    "winRate": 67,
    "starsDisplay": "15.9k",
    "weeklyStars": 7038,
    "weeklyGrowth": "+69.7%",
    "monthlyGrowth": "+22.4%",
    "wGrowthNum": 5.2,
    "mGrowthNum": 22.4,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/explodinggradients.png",
    "githubUrl": "https://github.com/explodinggradients/ragas"
  },
  {
    "id": "pedalboard",
    "name": "Spotify Pedalboard",
    "repo": "spotify/pedalboard",
    "author": "spotify",
    "persona": "Music & Audio Creators",
    "targetRole": "Sound Designers & Producers",
    "stars": 6339,
    "wGrowth": 40.9,
    "mGrowth": 22.4,
    "website": "https://github.com/spotify/pedalboard",
    "traffic": "320k/mo",
    "trafficNum": 320000,
    "searchVol": "42,000/mo",
    "searchVolNum": 42000,
    "searchTrend": "+45%",
    "latency": 18,
    "tokens": 98,
    "desc": "Spotify's fast Python library for adding audio effects, VSTs, and DSP guitar pedal chains to audio.",
    "useCase": "Programmatically apply reverb, compression, pitch shifts, and delay chains to thousands of audio samples.",
    "arch": "C++, JUCE audio framework, Python bindings.",
    "tags": [
      "audio-dsp",
      "effects",
      "vst3",
      "spotify"
    ],
    "rank": 251,
    "elo": 1150,
    "eloCi": "±23",
    "winRate": 66,
    "starsDisplay": "6.3k",
    "weeklyStars": 2073,
    "weeklyGrowth": "+40.9%",
    "monthlyGrowth": "+22.4%",
    "wGrowthNum": 5.2,
    "mGrowthNum": 22.4,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/spotify.png",
    "githubUrl": "https://github.com/spotify/pedalboard"
  },
  {
    "id": "posthog",
    "name": "PostHog",
    "repo": "PostHog/posthog",
    "author": "PostHog",
    "persona": "Marketers & Growth",
    "targetRole": "Growth Marketers",
    "stars": 40168,
    "wGrowth": 58.1,
    "mGrowth": 21.8,
    "website": "https://posthog.com",
    "traffic": "3.8M/mo",
    "trafficNum": 3800000,
    "searchVol": "160,000/mo",
    "searchVolNum": 160000,
    "searchTrend": "+38%",
    "latency": 48,
    "tokens": 89,
    "desc": "The all-in-one product analytics platform with session replay, feature flags, A/B testing, user surveys, and conversion funnels.",
    "useCase": "Pinpoint where users drop off in onboarding, watch anonymous session replays of stuck buttons, and trigger targeted promos.",
    "arch": "Python Django, ClickHouse, Kafka, React, Tailwind CSS.",
    "tags": [
      "product-analytics",
      "session-replay",
      "funnels",
      "ab-testing"
    ],
    "rank": 252,
    "elo": 1190,
    "eloCi": "±24",
    "winRate": 72,
    "starsDisplay": "40.2k",
    "weeklyStars": 16063,
    "weeklyGrowth": "+58.1%",
    "monthlyGrowth": "+21.8%",
    "wGrowthNum": 5.1,
    "mGrowthNum": 21.8,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/PostHog.png",
    "githubUrl": "https://github.com/PostHog/posthog"
  },
  {
    "id": "lean4",
    "name": "Lean 4",
    "repo": "leanprover/lean4",
    "author": "leanprover",
    "persona": "Researchers & STEM",
    "targetRole": "Pure Mathematicians",
    "stars": 9419,
    "wGrowth": 27.3,
    "mGrowth": 22,
    "website": "https://lean-lang.org",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "42,000/mo",
    "searchVolNum": 42000,
    "searchTrend": "+45%",
    "latency": 90,
    "tokens": 98,
    "desc": "Interactive theorem prover and functional programming language.",
    "useCase": "Formally verify Olympiad math solutions and software correctness proofs.",
    "arch": "C++, Lean kernel, dependent type theory.",
    "tags": [
      "formal-verification",
      "theorem-prover",
      "math"
    ],
    "rank": 253,
    "elo": 1154,
    "eloCi": "±25",
    "winRate": 67,
    "starsDisplay": "9.4k",
    "weeklyStars": 2396,
    "weeklyGrowth": "+27.3%",
    "monthlyGrowth": "+22.0%",
    "wGrowthNum": 5.1,
    "mGrowthNum": 22,
    "latencyMs": "90ms",
    "latencyNum": 90,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/leanprover.png",
    "githubUrl": "https://github.com/leanprover/lean4"
  },
  {
    "id": "ruff",
    "name": "Ruff",
    "repo": "astral-sh/ruff",
    "author": "astral-sh",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Python DevOps",
    "stars": 49931,
    "wGrowth": 37.9,
    "mGrowth": 20.6,
    "website": "https://astral.sh/ruff",
    "traffic": "1.5M/mo",
    "trafficNum": 1500000,
    "searchVol": "125,000/mo",
    "searchVolNum": 125000,
    "searchTrend": "+42%",
    "latency": 14,
    "tokens": 99,
    "desc": "An extremely fast Python linter and code formatter, written in Rust.",
    "useCase": "Replace Flake8, Black, isort, and Bandit with 100x faster single CLI execution.",
    "arch": "Rust, zero-overhead AST parser.",
    "tags": [
      "linter",
      "formatter",
      "python",
      "fast"
    ],
    "rank": 254,
    "elo": 1210,
    "eloCi": "±26",
    "winRate": 75,
    "starsDisplay": "49.9k",
    "weeklyStars": 15468,
    "weeklyGrowth": "+37.9%",
    "monthlyGrowth": "+20.6%",
    "wGrowthNum": 4.8,
    "mGrowthNum": 20.6,
    "latencyMs": "14ms",
    "latencyNum": 14,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "logoUrl": "https://github.com/astral-sh.png",
    "githubUrl": "https://github.com/astral-sh/ruff"
  },
  {
    "id": "qdrant",
    "name": "Qdrant Rust Vector DB",
    "repo": "qdrant/qdrant",
    "author": "qdrant",
    "persona": "AI Ops & Tooling",
    "targetRole": "RAG & Vector Search Engineers",
    "stars": 34957,
    "wGrowth": 50.7,
    "mGrowth": 20,
    "website": "https://qdrant.tech",
    "traffic": "980k/mo",
    "trafficNum": 980000,
    "searchVol": "62,000/mo",
    "searchVolNum": 62000,
    "searchTrend": "+60%",
    "latency": 12,
    "tokens": 94,
    "desc": "Vector Similarity Search Engine and Vector Database with extended payload-based filtering written in Rust.",
    "useCase": "Store dense embeddings for millions of enterprise documents with sub-10ms query latency and boolean payload filtering.",
    "arch": "Rust, HNSW indexing, gRPC API, Raft consensus, Memory-mapped files.",
    "tags": [
      "vector-db",
      "rust",
      "hnsw",
      "rag",
      "similarity-search"
    ],
    "rank": 255,
    "elo": 1184,
    "eloCi": "±18",
    "winRate": 71,
    "starsDisplay": "35.0k",
    "weeklyStars": 12870,
    "weeklyGrowth": "+50.7%",
    "monthlyGrowth": "+20.0%",
    "wGrowthNum": 4.8,
    "mGrowthNum": 20,
    "latencyMs": "12ms",
    "latencyNum": 12,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/qdrant.png",
    "githubUrl": "https://github.com/qdrant/qdrant"
  },
  {
    "id": "biome",
    "name": "Biome",
    "repo": "biomejs/biome",
    "author": "biomejs",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Frontend Tooling Engineers",
    "stars": 25909,
    "wGrowth": 57,
    "mGrowth": 20.5,
    "website": "https://biomejs.dev",
    "traffic": "780k/mo",
    "trafficNum": 780000,
    "searchVol": "68,000/mo",
    "searchVolNum": 68000,
    "searchTrend": "+42%",
    "latency": 15,
    "tokens": 99,
    "desc": "One toolchain for your web project. Formats, lints, and parses code in sub-10ms.",
    "useCase": "Run instant lint and formatting verification in agentic git commit loops.",
    "arch": "Rust, multi-threaded AST, sub-millisecond CLI.",
    "tags": [
      "rust",
      "formatter",
      "linter",
      "toolchain"
    ],
    "rank": 256,
    "elo": 1171,
    "eloCi": "±19",
    "winRate": 69,
    "starsDisplay": "25.9k",
    "weeklyStars": 10201,
    "weeklyGrowth": "+57%",
    "monthlyGrowth": "+20.5%",
    "wGrowthNum": 4.8,
    "mGrowthNum": 20.5,
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "logoUrl": "https://github.com/biomejs.png",
    "githubUrl": "https://github.com/biomejs/biome"
  },
  {
    "id": "audiocraft",
    "name": "AudioCraft (MusicGen)",
    "repo": "facebookresearch/audiocraft",
    "author": "facebookresearch",
    "persona": "Music & Audio Creators",
    "targetRole": "Composers & Sound Designers",
    "stars": 23661,
    "wGrowth": 4.8,
    "mGrowth": 20.8,
    "website": "https://audiocraft.metademolab.com",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "185,000/mo",
    "searchVolNum": 185000,
    "searchTrend": "+36%",
    "latency": 220,
    "tokens": 85,
    "desc": "Meta AI's open generative audio library featuring MusicGen (text-to-music) and AudioGen (text-to-sound-effects) autoregressive models.",
    "useCase": "Generate bespoke background scores, ambient soundscapes, and instrumental melodies from natural language style descriptions.",
    "arch": "EnCodec neural audio codec, Transformer LM, PyTorch, Gradio.",
    "tags": [
      "text-to-music",
      "musicgen",
      "generative-audio",
      "soundtrack"
    ],
    "rank": 257,
    "elo": 1183,
    "eloCi": "±20",
    "winRate": 71,
    "starsDisplay": "23.7k",
    "weeklyStars": 1955,
    "weeklyGrowth": "+4.8%",
    "monthlyGrowth": "+20.8%",
    "wGrowthNum": 4.8,
    "mGrowthNum": 20.8,
    "latencyMs": "220ms",
    "latencyNum": 220,
    "tokensSaved": "-85%",
    "tokensNum": 85,
    "logoUrl": "https://github.com/facebookresearch.png",
    "githubUrl": "https://github.com/facebookresearch/audiocraft"
  },
  {
    "id": "n8n",
    "name": "n8n",
    "repo": "n8n-io/n8n",
    "author": "n8n-io",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Ops & Technical Founders",
    "stars": 206788,
    "wGrowth": 255.3,
    "mGrowth": 19.5,
    "website": "https://n8n.io",
    "traffic": "3.4M/mo",
    "trafficNum": 3400000,
    "searchVol": "380,000/mo",
    "searchVolNum": 380000,
    "searchTrend": "+35%",
    "latency": 60,
    "tokens": 84,
    "desc": "Fair-code workflow automation platform with AI agent nodes, custom code execution, and multi-step branching logic.",
    "useCase": "Orchestrate complex customer lead routing, automated trial extension logic, and multi-tenant webhook listeners with visual retry logic.",
    "arch": "TypeScript, Vue.js, SQLite/PostgreSQL, LangChain integration.",
    "tags": [
      "workflow-automation",
      "agentic-nodes",
      "backend-ops"
    ],
    "rank": 258,
    "elo": 1252,
    "eloCi": "±21",
    "winRate": 81,
    "starsDisplay": "206.8k",
    "weeklyStars": 151207,
    "weeklyGrowth": "+255.3%",
    "monthlyGrowth": "+19.5%",
    "wGrowthNum": 4.5,
    "mGrowthNum": 19.5,
    "latencyMs": "60ms",
    "latencyNum": 60,
    "tokensSaved": "-84%",
    "tokensNum": 84,
    "logoUrl": "https://github.com/n8n-io.png",
    "githubUrl": "https://github.com/n8n-io/n8n",
    "todayBreakthrough": "AI agent"
  },
  {
    "id": "swe-bench",
    "name": "SWE-bench",
    "repo": "swe-bench/SWE-bench",
    "author": "swe-bench",
    "persona": "Autonomous Software Engineers",
    "targetRole": "AI Evaluation Engineers",
    "stars": 5985,
    "wGrowth": 4.5,
    "mGrowth": 19.5,
    "website": "https://swebench.com",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "42,000/mo",
    "searchVolNum": 42000,
    "searchTrend": "+38%",
    "latency": 60,
    "tokens": 94,
    "desc": "The gold standard evaluation benchmark testing AI systems on resolving end-to-end software engineering problems from real GitHub repos.",
    "useCase": "Evaluate whether your custom agent prompts actually fix complex Python bugs or introduce hidden regressions.",
    "arch": "Docker evaluation harness, PyTest, Git patch validator.",
    "tags": [
      "evaluation",
      "benchmark",
      "swe-gold-standard",
      "accuracy"
    ],
    "rank": 259,
    "elo": 1147,
    "eloCi": "±22",
    "winRate": 66,
    "starsDisplay": "6.0k",
    "weeklyStars": 446,
    "weeklyGrowth": "+4.5%",
    "monthlyGrowth": "+19.5%",
    "wGrowthNum": 4.5,
    "mGrowthNum": 19.5,
    "latencyMs": "60ms",
    "latencyNum": 60,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/swe-bench.png",
    "githubUrl": "https://github.com/swe-bench/SWE-bench"
  },
  {
    "id": "whisper-cpp",
    "name": "Whisper.cpp",
    "repo": "ggerganov/whisper.cpp",
    "author": "ggerganov",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Engineers & Podcasters",
    "stars": 54186,
    "wGrowth": 47.2,
    "mGrowth": 18,
    "website": "https://github.com/ggerganov/whisper.cpp",
    "traffic": "920k/mo",
    "trafficNum": 920000,
    "searchVol": "74,000/mo",
    "searchVolNum": 74000,
    "searchTrend": "+42%",
    "latency": 18,
    "tokens": 95,
    "desc": "High-performance inference of OpenAI Whisper automatic speech recognition in pure C/C++ with zero dependencies.",
    "useCase": "Transcribe podcast stems, generate synchronized lyric captions (SRT/VTT), and perform real-time voice command decoding on Apple Silicon and WASM.",
    "arch": "Pure C/C++, GGML tensor library, Metal/CUDA acceleration, WebAssembly exports.",
    "tags": [
      "speech-to-text",
      "audio-transcription",
      "c-plus-plus",
      "whisper",
      "ggml"
    ],
    "rank": 260,
    "elo": 1206,
    "eloCi": "±24",
    "winRate": 74,
    "starsDisplay": "54.2k",
    "weeklyStars": 18931,
    "weeklyGrowth": "+47.2%",
    "monthlyGrowth": "+18.0%",
    "wGrowthNum": 4.2,
    "mGrowthNum": 18,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/ggerganov.png",
    "githubUrl": "https://github.com/ggerganov/whisper.cpp"
  },
  {
    "id": "umami",
    "name": "Umami",
    "repo": "umami-software/umami",
    "author": "umami-software",
    "persona": "Marketers & Growth",
    "targetRole": "Privacy-Conscious Marketers",
    "stars": 39214,
    "wGrowth": 62.7,
    "mGrowth": 18,
    "website": "https://umami.is",
    "traffic": "980k/mo",
    "trafficNum": 980000,
    "searchVol": "62,000/mo",
    "searchVolNum": 62000,
    "searchTrend": "+31%",
    "latency": 30,
    "tokens": 94,
    "desc": "Umami is an open source, privacy-focused alternative to Google Analytics.",
    "useCase": "Track referral conversions and organic landing page visits without cookies.",
    "arch": "Next.js, Prisma, PostgreSQL/MySQL.",
    "tags": [
      "analytics",
      "privacy",
      "gdpr",
      "metrics"
    ],
    "rank": 261,
    "elo": 1181,
    "eloCi": "±25",
    "winRate": 70,
    "starsDisplay": "39.2k",
    "weeklyStars": 16126,
    "weeklyGrowth": "+62.7%",
    "monthlyGrowth": "+18.0%",
    "wGrowthNum": 4.2,
    "mGrowthNum": 18,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/umami-software.png",
    "githubUrl": "https://github.com/umami-software/umami"
  },
  {
    "id": "react-email",
    "name": "React Email",
    "repo": "resend/react-email",
    "author": "resend",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Product Designers & Founders",
    "stars": 19814,
    "wGrowth": 33.9,
    "mGrowth": 18,
    "website": "https://react.email",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "38,000/mo",
    "searchVolNum": 38000,
    "searchTrend": "+28%",
    "latency": 28,
    "tokens": 95,
    "desc": "Collection of high-taste, responsive email components written in React. Generates cross-client compatible HTML for Apple Mail, Gmail, and Outlook.",
    "useCase": "Build bespoke transactional onboarding sequences, invoice receipts, and password resets that never render broken on mobile mail clients.",
    "arch": "React 19, Tailwind CSS, TypeScript, Server-Side HTML Compiler.",
    "tags": [
      "transactional-email",
      "resend",
      "newsletter",
      "onboarding"
    ],
    "rank": 262,
    "elo": 1162,
    "eloCi": "±26",
    "winRate": 68,
    "starsDisplay": "19.8k",
    "weeklyStars": 5635,
    "weeklyGrowth": "+33.9%",
    "monthlyGrowth": "+18.0%",
    "wGrowthNum": 4.2,
    "mGrowthNum": 18,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/resend.png",
    "githubUrl": "https://github.com/resend/react-email"
  },
  {
    "id": "medusa",
    "name": "Medusa",
    "repo": "medusajs/medusa",
    "author": "medusajs",
    "persona": "Founders & Indie Hackers",
    "targetRole": "E-Commerce Founders",
    "stars": 36633,
    "wGrowth": 32.7,
    "mGrowth": 17.5,
    "website": "https://medusajs.com",
    "traffic": "890k/mo",
    "trafficNum": 890000,
    "searchVol": "88,000/mo",
    "searchVolNum": 88000,
    "searchTrend": "+26%",
    "latency": 48,
    "tokens": 89,
    "desc": "The open source Shopify alternative for bespoke modular commerce backends.",
    "useCase": "Build custom multi-currency checkout funnels without Shopify revenue share.",
    "arch": "Node.js, PostgreSQL, Redis, Medusa Workflows.",
    "tags": [
      "ecommerce",
      "shopify-alt",
      "checkout"
    ],
    "rank": 263,
    "elo": 1187,
    "eloCi": "±18",
    "winRate": 71,
    "starsDisplay": "36.6k",
    "weeklyStars": 10164,
    "weeklyGrowth": "+32.7%",
    "monthlyGrowth": "+17.5%",
    "wGrowthNum": 4.1,
    "mGrowthNum": 17.5,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/medusajs.png",
    "githubUrl": "https://github.com/medusajs/medusa"
  },
  {
    "id": "chroma",
    "name": "Chroma Embedding Store",
    "repo": "chroma-core/chroma",
    "author": "chroma-core",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Local AI & RAG Engineers",
    "stars": 29457,
    "wGrowth": 60.1,
    "mGrowth": 17.5,
    "website": "https://trychroma.com",
    "traffic": "740k/mo",
    "trafficNum": 740000,
    "searchVol": "58,000/mo",
    "searchVolNum": 58000,
    "searchTrend": "+45%",
    "latency": 20,
    "tokens": 92,
    "desc": "The AI-native open-source embedding database that gives LLMs memory, document retrieval, and semantic search.",
    "useCase": "Embed local codebases and GitHub PR diffs into ephemeral vector stores for instantaneous agent recall during development.",
    "arch": "Python, TypeScript, ClickHouse/DuckDB backing, HNSWLib indexer.",
    "tags": [
      "embedding-db",
      "llm-memory",
      "semantic-search",
      "zero-config"
    ],
    "rank": 264,
    "elo": 1168,
    "eloCi": "±20",
    "winRate": 69,
    "starsDisplay": "29.5k",
    "weeklyStars": 11811,
    "weeklyGrowth": "+60.1%",
    "monthlyGrowth": "+17.5%",
    "wGrowthNum": 4.1,
    "mGrowthNum": 17.5,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/chroma-core.png",
    "githubUrl": "https://github.com/chroma-core/chroma"
  },
  {
    "id": "novuhq",
    "name": "Novu",
    "repo": "novuhq/novu",
    "author": "novuhq",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Full-Stack Devs & Founders",
    "stars": 40120,
    "wGrowth": 14.3,
    "mGrowth": 17,
    "website": "https://novu.co",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "48,000/mo",
    "searchVolNum": 48000,
    "searchTrend": "+28%",
    "latency": 40,
    "tokens": 91,
    "desc": "The open-source notification infrastructure for developers (Email, In-App, Push, Chat).",
    "useCase": "Unify user notification delivery across Slack, SMS, and transactional email.",
    "arch": "Node.js, NestJS, Redis, MongoDB.",
    "tags": [
      "notifications",
      "email",
      "sms",
      "push"
    ],
    "rank": 265,
    "elo": 1202,
    "eloCi": "±21",
    "winRate": 73,
    "starsDisplay": "40.1k",
    "weeklyStars": 6424,
    "weeklyGrowth": "+14.3%",
    "monthlyGrowth": "+17.0%",
    "wGrowthNum": 4,
    "mGrowthNum": 17,
    "latencyMs": "40ms",
    "latencyNum": 40,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/novuhq.png",
    "githubUrl": "https://github.com/novuhq/novu"
  },
  {
    "id": "cal-com",
    "name": "Cal.com Scheduling",
    "repo": "calcom/cal.com",
    "author": "calcom",
    "persona": "Founders & Indie Hackers",
    "targetRole": "B2B Founders & Sales",
    "stars": 48903,
    "wGrowth": 41.7,
    "mGrowth": 16.5,
    "website": "https://cal.com",
    "traffic": "2.6M/mo",
    "trafficNum": 2600000,
    "searchVol": "190,000/mo",
    "searchVolNum": 190000,
    "searchTrend": "+32%",
    "latency": 42,
    "tokens": 89,
    "desc": "Self-hostable, extensible scheduling infrastructure replacing Calendly with white-label booking, CRM sync, and Stripe deposits.",
    "useCase": "Embed demo booking widgets on landing pages that qualify inbound B2B leads before placing calls on founder calendars.",
    "arch": "Next.js, Prisma, PostgreSQL, tRPC, Tailwind CSS, Stripe webhook listeners.",
    "tags": [
      "scheduling",
      "calendly-alt",
      "crm",
      "lead-capture"
    ],
    "rank": 266,
    "elo": 1200,
    "eloCi": "±22",
    "winRate": 73,
    "starsDisplay": "48.9k",
    "weeklyStars": 15748,
    "weeklyGrowth": "+41.7%",
    "monthlyGrowth": "+16.5%",
    "wGrowthNum": 3.9,
    "mGrowthNum": 16.5,
    "latencyMs": "42ms",
    "latencyNum": 42,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/calcom.png",
    "githubUrl": "https://github.com/calcom/cal.com"
  },
  {
    "id": "supabase",
    "name": "Supabase",
    "repo": "supabase/supabase",
    "author": "supabase",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Full-Stack Founders",
    "stars": 111188,
    "wGrowth": 45.5,
    "mGrowth": 16.2,
    "website": "https://supabase.com",
    "traffic": "4.8M/mo",
    "trafficNum": 4800000,
    "searchVol": "450,000/mo",
    "searchVolNum": 450000,
    "searchTrend": "+28%",
    "latency": 28,
    "tokens": 92,
    "desc": "Open-source Firebase alternative providing Postgres database, Authentication, instant Realtime subscriptions, Edge Functions, and Storage in minutes.",
    "useCase": "Ship production SaaS backends in 1 day with relational SQL, row-level security, and Stripe webhook handling.",
    "arch": "PostgreSQL, PostgREST, GoTrue Auth, Realtime Elixir engine, Kong API Gateway.",
    "tags": [
      "postgres",
      "auth",
      "realtime",
      "backend",
      "indie-hacker"
    ],
    "rank": 267,
    "elo": 1282,
    "eloCi": "±23",
    "winRate": 85,
    "starsDisplay": "111.2k",
    "weeklyStars": 37691,
    "weeklyGrowth": "+45.5%",
    "monthlyGrowth": "+16.2%",
    "wGrowthNum": 3.8,
    "mGrowthNum": 16.2,
    "latencyMs": "28ms",
    "latencyNum": 28,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/supabase.png",
    "githubUrl": "https://github.com/supabase/supabase"
  },
  {
    "id": "open-interpreter",
    "name": "Open Interpreter",
    "repo": "OpenInterpreter/open-interpreter",
    "author": "OpenInterpreter",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Terminal Hackers",
    "stars": 68520,
    "wGrowth": 20.6,
    "mGrowth": 16,
    "website": "https://openinterpreter.com",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "180,000/mo",
    "searchVolNum": 180000,
    "searchTrend": "+28%",
    "latency": 65,
    "tokens": 88,
    "desc": "A natural language interface for computers to run code in Python, Bash, and AppleScript.",
    "useCase": "Automate OS-level file manipulation, browser scraping, and script debugging.",
    "arch": "Python, Jupyter kernel, LiteLLM.",
    "tags": [
      "terminal-os",
      "voice-coding",
      "agent"
    ],
    "rank": 268,
    "elo": 1243,
    "eloCi": "±24",
    "winRate": 79,
    "starsDisplay": "68.5k",
    "weeklyStars": 13878,
    "weeklyGrowth": "+20.6%",
    "monthlyGrowth": "+16.0%",
    "wGrowthNum": 3.8,
    "mGrowthNum": 16,
    "latencyMs": "65ms",
    "latencyNum": 65,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/OpenInterpreter.png",
    "githubUrl": "https://github.com/OpenInterpreter/open-interpreter"
  },
  {
    "id": "rsshub",
    "name": "RSSHub",
    "repo": "DIYgod/RSSHub",
    "author": "DIYgod",
    "persona": "Marketers & Growth",
    "targetRole": "Market Intelligence Leads",
    "stars": 46431,
    "wGrowth": 34.6,
    "mGrowth": 16.5,
    "website": "https://rsshub.app",
    "traffic": "1.8M/mo",
    "trafficNum": 1800000,
    "searchVol": "92,000/mo",
    "searchVolNum": 92000,
    "searchTrend": "+28%",
    "latency": 38,
    "tokens": 91,
    "desc": "Everything is RSS-able! Extensible radar generating clean RSS feeds from any website, social platform, or walled app.",
    "useCase": "Monitor competitor press releases, product updates, brand mentions, and industry discussions across 50 platforms in real-time.",
    "arch": "Node.js, Cheerio, Puppeteer, Redis, Docker.",
    "tags": [
      "market-signals",
      "rss",
      "social-monitoring",
      "competitor-radar"
    ],
    "rank": 269,
    "elo": 1199,
    "eloCi": "±25",
    "winRate": 73,
    "starsDisplay": "46.4k",
    "weeklyStars": 13242,
    "weeklyGrowth": "+34.6%",
    "monthlyGrowth": "+16.5%",
    "wGrowthNum": 3.8,
    "mGrowthNum": 16.5,
    "latencyMs": "38ms",
    "latencyNum": 38,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/DIYgod.png",
    "githubUrl": "https://github.com/DIYgod/RSSHub"
  },
  {
    "id": "pocketbase",
    "name": "PocketBase",
    "repo": "pocketbase/pocketbase",
    "author": "pocketbase",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Micro-SaaS Founders",
    "stars": 61306,
    "wGrowth": 40,
    "mGrowth": 15,
    "website": "https://pocketbase.io",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+25%",
    "latency": 18,
    "tokens": 98,
    "desc": "Open source backend consisting of embedded SQLite database with real-time subscriptions, built-in files and auth management, and admin dashboard.",
    "useCase": "Run a complete SaaS backend for $4/mo on a tiny VPS that easily handles 10,000 daily active users with sub-10ms response times.",
    "arch": "Go single binary, SQLite + WAL mode, Echo framework.",
    "tags": [
      "sqlite",
      "single-binary",
      "zero-ops",
      "micro-saas"
    ],
    "rank": 270,
    "elo": 1215,
    "eloCi": "±26",
    "winRate": 75,
    "starsDisplay": "61.3k",
    "weeklyStars": 19039,
    "weeklyGrowth": "+40%",
    "monthlyGrowth": "+15.0%",
    "wGrowthNum": 3.5,
    "mGrowthNum": 15,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/pocketbase.png",
    "githubUrl": "https://github.com/pocketbase/pocketbase"
  },
  {
    "id": "coqui-tts",
    "name": "Coqui TTS",
    "repo": "coqui-ai/TTS",
    "author": "coqui-ai",
    "persona": "Music & Audio Creators",
    "targetRole": "Voiceover Artists & Creators",
    "stars": 46101,
    "wGrowth": 34.8,
    "mGrowth": 15.2,
    "website": "https://github.com/coqui-ai/TTS",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "160,000/mo",
    "searchVolNum": 160000,
    "searchTrend": "+18%",
    "latency": 90,
    "tokens": 90,
    "desc": "Deep learning toolkit for Text-to-Speech with voice cloning, multi-speaker models (XTTS v2), and real-time streaming.",
    "useCase": "Produce studio-grade narration, audiobooks, and emotional voiceovers from 3-second reference audio clips.",
    "arch": "XTTS, VITS, Tacotron2, PyTorch, ONNX runtime.",
    "tags": [
      "voice-cloning",
      "tts",
      "speech-synthesis",
      "narration"
    ],
    "rank": 271,
    "elo": 1196,
    "eloCi": "±18",
    "winRate": 73,
    "starsDisplay": "46.1k",
    "weeklyStars": 13098,
    "weeklyGrowth": "+34.8%",
    "monthlyGrowth": "+15.2%",
    "wGrowthNum": 3.5,
    "mGrowthNum": 15.2,
    "latencyMs": "90ms",
    "latencyNum": 90,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/coqui-ai.png",
    "githubUrl": "https://github.com/coqui-ai/TTS"
  },
  {
    "id": "invokeai",
    "name": "InvokeAI Creative Engine",
    "repo": "invoke-ai/InvokeAI",
    "author": "invoke-ai",
    "persona": "Music & Audio Creators",
    "targetRole": "Creative Directors & Designers",
    "stars": 28393,
    "wGrowth": 21.3,
    "mGrowth": 15,
    "website": "https://invoke.com",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "32,000/mo",
    "searchVolNum": 32000,
    "searchTrend": "+25%",
    "latency": 85,
    "tokens": 91,
    "desc": "Professional creative engine and infinite canvas for visual generative AI, model training, and asset pipelines.",
    "useCase": "Design continuous high-resolution visual boards and canvas storyboards for song trailers and product launch videos.",
    "arch": "Python, React, TypeScript, FastAPI, PyTorch, Diffusers.",
    "tags": [
      "canvas-ui",
      "creative-suite",
      "diffusers",
      "visual-storyboarding"
    ],
    "rank": 272,
    "elo": 1174,
    "eloCi": "±19",
    "winRate": 69,
    "starsDisplay": "28.4k",
    "weeklyStars": 5812,
    "weeklyGrowth": "+21.3%",
    "monthlyGrowth": "+15.0%",
    "wGrowthNum": 3.5,
    "mGrowthNum": 15,
    "latencyMs": "85ms",
    "latencyNum": 85,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/invoke-ai.png",
    "githubUrl": "https://github.com/invoke-ai/InvokeAI"
  },
  {
    "id": "chatwoot",
    "name": "Chatwoot",
    "repo": "chatwoot/chatwoot",
    "author": "chatwoot",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Customer Success & Founders",
    "stars": 37598,
    "wGrowth": 74.9,
    "mGrowth": 15,
    "website": "https://chatwoot.com",
    "traffic": "950k/mo",
    "trafficNum": 950000,
    "searchVol": "52,000/mo",
    "searchVolNum": 52000,
    "searchTrend": "+22%",
    "latency": 45,
    "tokens": 89,
    "desc": "Customer communication platform. An alternative to Intercom, Zendesk, and Drift.",
    "useCase": "Embed real-time customer chat with live agent routing directly in your SaaS.",
    "arch": "Ruby on Rails, Vue.js, PostgreSQL, Redis.",
    "tags": [
      "customer-support",
      "live-chat",
      "intercom-alt"
    ],
    "rank": 273,
    "elo": 1171,
    "eloCi": "±20",
    "winRate": 69,
    "starsDisplay": "37.6k",
    "weeklyStars": 16850,
    "weeklyGrowth": "+74.9%",
    "monthlyGrowth": "+15.0%",
    "wGrowthNum": 3.5,
    "mGrowthNum": 15,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/chatwoot.png",
    "githubUrl": "https://github.com/chatwoot/chatwoot"
  },
  {
    "id": "shadcn-ui",
    "name": "shadcn/ui",
    "repo": "shadcn-ui/ui",
    "author": "shadcn-ui",
    "persona": "Founders & Indie Hackers",
    "targetRole": "UI/UX Designers & Devs",
    "stars": 125231,
    "wGrowth": 63.9,
    "mGrowth": 14.8,
    "website": "https://ui.shadcn.com",
    "traffic": "5.4M/mo",
    "trafficNum": 5400000,
    "searchVol": "280,000/mo",
    "searchVolNum": 280000,
    "searchTrend": "+35%",
    "latency": 25,
    "tokens": 96,
    "desc": "Beautifully designed, accessible components that you copy and paste into your apps. Accessible, customizable, and open source.",
    "useCase": "Create modern SaaS user interfaces, metrics dashboards, and checkout forms without heavy external component libraries.",
    "arch": "Radix UI Primitives, Tailwind CSS, Class Variance Authority, React.",
    "tags": [
      "components",
      "radix",
      "tailwind",
      "design-system"
    ],
    "rank": 274,
    "elo": 1279,
    "eloCi": "±22",
    "winRate": 84,
    "starsDisplay": "125.2k",
    "weeklyStars": 51428,
    "weeklyGrowth": "+63.9%",
    "monthlyGrowth": "+14.8%",
    "wGrowthNum": 3.4,
    "mGrowthNum": 14.8,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/shadcn-ui.png",
    "githubUrl": "https://github.com/shadcn-ui/ui"
  },
  {
    "id": "cleanlab",
    "name": "Cleanlab",
    "repo": "cleanlab/cleanlab",
    "author": "cleanlab",
    "persona": "Researchers & STEM",
    "targetRole": "Data-Centric AI Scientists",
    "stars": 11689,
    "wGrowth": 27.1,
    "mGrowth": 14.6,
    "website": "https://cleanlab.ai",
    "traffic": "410k/mo",
    "trafficNum": 410000,
    "searchVol": "31,000/mo",
    "searchVolNum": 31000,
    "searchTrend": "+25%",
    "latency": 32,
    "tokens": 96,
    "desc": "The standard open-source library for data-centric AI and label quality improvement.",
    "useCase": "Detect noisy labels and corrupted training tokens in multimodal datasets.",
    "arch": "Python, scikit-learn, PyTorch.",
    "tags": [
      "data-centric",
      "label-errors",
      "dataset-hygiene"
    ],
    "rank": 275,
    "elo": 1145,
    "eloCi": "±23",
    "winRate": 65,
    "starsDisplay": "11.7k",
    "weeklyStars": 2801,
    "weeklyGrowth": "+27.1%",
    "monthlyGrowth": "+14.6%",
    "wGrowthNum": 3.4,
    "mGrowthNum": 14.6,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/cleanlab.png",
    "githubUrl": "https://github.com/cleanlab/cleanlab"
  },
  {
    "id": "whisper",
    "name": "Whisper",
    "repo": "openai/whisper",
    "author": "openai",
    "persona": "Music & Audio Creators",
    "targetRole": "Podcasters & Video Creators",
    "stars": 110072,
    "wGrowth": 43.9,
    "mGrowth": 13.8,
    "website": "https://openai.com/research/whisper",
    "traffic": "8.5M/mo",
    "trafficNum": 8500000,
    "searchVol": "580,000/mo",
    "searchVolNum": 580000,
    "searchTrend": "+25%",
    "latency": 75,
    "tokens": 94,
    "desc": "Robust speech recognition model trained on 680,000 hours of multilingual audio. Transcribes and translates speech into timestamped text.",
    "useCase": "Generate exact lyric timestamps for lyric videos, podcast show notes, and automated video subtitle burning.",
    "arch": "Encoder-Decoder Transformer, PyTorch, Mel-Spectrogram Front-end.",
    "tags": [
      "transcription",
      "subtitles",
      "lyrics",
      "audio-to-text"
    ],
    "rank": 276,
    "elo": 1278,
    "eloCi": "±24",
    "winRate": 84,
    "starsDisplay": "110.1k",
    "weeklyStars": 36020,
    "weeklyGrowth": "+43.9%",
    "monthlyGrowth": "+13.8%",
    "wGrowthNum": 3.2,
    "mGrowthNum": 13.8,
    "latencyMs": "75ms",
    "latencyNum": 75,
    "tokensSaved": "-94%",
    "tokensNum": 94,
    "logoUrl": "https://github.com/openai.png",
    "githubUrl": "https://github.com/openai/whisper"
  },
  {
    "id": "milvus",
    "name": "Milvus Distributed Vector DB",
    "repo": "milvus-io/milvus",
    "author": "milvus-io",
    "persona": "AI Ops & Tooling",
    "targetRole": "Enterprise Infrastructure Engineers",
    "stars": 46329,
    "wGrowth": 43,
    "mGrowth": 13.5,
    "website": "https://milvus.io",
    "traffic": "810k/mo",
    "trafficNum": 810000,
    "searchVol": "70,000/mo",
    "searchVolNum": 70000,
    "searchTrend": "+35%",
    "latency": 15,
    "tokens": 95,
    "desc": "A distributed vector database built for scalable similarity search and massive AI workloads.",
    "useCase": "Index billion-scale vector collections across Kubernetes clusters for enterprise search and multi-modal recommendation systems.",
    "arch": "Go, C++, Knowhere index core, Kafka/Pulsar log broker, MinIO object store.",
    "tags": [
      "distributed-vector-db",
      "billion-scale",
      "kubernetes",
      "enterprise"
    ],
    "rank": 277,
    "elo": 1189,
    "eloCi": "±25",
    "winRate": 72,
    "starsDisplay": "46.3k",
    "weeklyStars": 14965,
    "weeklyGrowth": "+43%",
    "monthlyGrowth": "+13.5%",
    "wGrowthNum": 3.2,
    "mGrowthNum": 13.5,
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/milvus-io.png",
    "githubUrl": "https://github.com/milvus-io/milvus"
  },
  {
    "id": "typesense",
    "name": "Typesense",
    "repo": "typesense/typesense",
    "author": "typesense",
    "persona": "Marketers & Growth",
    "targetRole": "Product Search Devs",
    "stars": 26637,
    "wGrowth": 30.6,
    "mGrowth": 13.8,
    "website": "https://typesense.org",
    "traffic": "890k/mo",
    "trafficNum": 890000,
    "searchVol": "58,000/mo",
    "searchVolNum": 58000,
    "searchTrend": "+21%",
    "latency": 18,
    "tokens": 97,
    "desc": "Fast, typo-tolerant search engine built in C++ for instantaneous search experiences.",
    "useCase": "In-memory instant faceted filtering for e-commerce directories.",
    "arch": "C++, Raft consensus, in-memory index.",
    "tags": [
      "search",
      "algolia-alt",
      "in-memory"
    ],
    "rank": 278,
    "elo": 1165,
    "eloCi": "±26",
    "winRate": 68,
    "starsDisplay": "26.6k",
    "weeklyStars": 6889,
    "weeklyGrowth": "+30.6%",
    "monthlyGrowth": "+13.8%",
    "wGrowthNum": 3.2,
    "mGrowthNum": 13.8,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/typesense.png",
    "githubUrl": "https://github.com/typesense/typesense"
  },
  {
    "id": "plausible",
    "name": "Plausible",
    "repo": "plausible/analytics",
    "author": "plausible",
    "persona": "Marketers & Growth",
    "targetRole": "SEO & Growth Leads",
    "stars": 29334,
    "wGrowth": 41,
    "mGrowth": 13.5,
    "website": "https://plausible.io",
    "traffic": "2.1M/mo",
    "trafficNum": 2100000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+20%",
    "latency": 22,
    "tokens": 95,
    "desc": "Lightweight (< 1 KB) open-source web analytics. No cookies, fully compliant with GDPR, CCPA and PECR out of the box.",
    "useCase": "Measure clean site visitor traffic, Google Search keywords, and referral sources without slowing down page load speed.",
    "arch": "Elixir OTP, Phoenix, ClickHouse, Tailwind CSS.",
    "tags": [
      "privacy-analytics",
      "gdpr-safe",
      "lightweight",
      "page-speed"
    ],
    "rank": 279,
    "elo": 1165,
    "eloCi": "±18",
    "winRate": 68,
    "starsDisplay": "29.3k",
    "weeklyStars": 9178,
    "weeklyGrowth": "+41%",
    "monthlyGrowth": "+13.5%",
    "wGrowthNum": 3.1,
    "mGrowthNum": 13.5,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/plausible.png",
    "githubUrl": "https://github.com/plausible/analytics"
  },
  {
    "id": "flagsmith",
    "name": "Flagsmith",
    "repo": "Flagsmith/flagsmith",
    "author": "Flagsmith",
    "persona": "Marketers & Growth",
    "targetRole": "Product & Growth Ops",
    "stars": 6590,
    "wGrowth": 26.7,
    "mGrowth": 13.3,
    "website": "https://flagsmith.com",
    "traffic": "380k/mo",
    "trafficNum": 380000,
    "searchVol": "25,000/mo",
    "searchVolNum": 25000,
    "searchTrend": "+20%",
    "latency": 35,
    "tokens": 93,
    "desc": "Feature flag and remote config service. Manage features across web, mobile, and server.",
    "useCase": "Gradually roll out new pricing plans to 10% of users to test conversion impact.",
    "arch": "Python Django, PostgreSQL, React.",
    "tags": [
      "feature-flags",
      "rollouts",
      "ab-testing"
    ],
    "rank": 280,
    "elo": 1134,
    "eloCi": "±19",
    "winRate": 64,
    "starsDisplay": "6.6k",
    "weeklyStars": 1551,
    "weeklyGrowth": "+26.7%",
    "monthlyGrowth": "+13.3%",
    "wGrowthNum": 3.1,
    "mGrowthNum": 13.3,
    "latencyMs": "35ms",
    "latencyNum": 35,
    "tokensSaved": "-93%",
    "tokensNum": 93,
    "logoUrl": "https://github.com/Flagsmith.png",
    "githubUrl": "https://github.com/Flagsmith/flagsmith"
  },
  {
    "id": "astro",
    "name": "Astro Web Framework",
    "repo": "withastro/astro",
    "author": "withastro",
    "persona": "Marketers & Growth",
    "targetRole": "SEO & Web Developers",
    "stars": 63085,
    "wGrowth": 28.2,
    "mGrowth": 12.9,
    "website": "https://astro.build",
    "traffic": "2.8M/mo",
    "trafficNum": 2800000,
    "searchVol": "280,000/mo",
    "searchVolNum": 280000,
    "searchTrend": "+20%",
    "latency": 22,
    "tokens": 98,
    "desc": "The web framework for content-driven websites. Zero JS by default with Islands architecture.",
    "useCase": "Ship marketing sites with perfect 100/100 Core Web Vitals and instant page rendering.",
    "arch": "Vite, Rollup, TypeScript, Islands Architecture.",
    "tags": [
      "seo",
      "web-framework",
      "cwv",
      "fast"
    ],
    "rank": 281,
    "elo": 1222,
    "eloCi": "±20",
    "winRate": 76,
    "starsDisplay": "63.1k",
    "weeklyStars": 15361,
    "weeklyGrowth": "+28.2%",
    "monthlyGrowth": "+12.9%",
    "wGrowthNum": 3,
    "mGrowthNum": 12.9,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/withastro.png",
    "githubUrl": "https://github.com/withastro/astro"
  },
  {
    "id": "meilisearch",
    "name": "Meilisearch",
    "repo": "meilisearch/meilisearch",
    "author": "meilisearch",
    "persona": "Marketers & Growth",
    "targetRole": "Search Experience Leads",
    "stars": 59505,
    "wGrowth": 24.5,
    "mGrowth": 12.4,
    "website": "https://meilisearch.com",
    "traffic": "1.5M/mo",
    "trafficNum": 1500000,
    "searchVol": "125,000/mo",
    "searchVolNum": 125000,
    "searchTrend": "+18%",
    "latency": 20,
    "tokens": 96,
    "desc": "Lightning fast, hyper-relevant search engine with typo-tolerance out of the box.",
    "useCase": "Provide sub-10ms instant search across 50,000 catalog items.",
    "arch": "Rust, LMDB, Actix-web.",
    "tags": [
      "search",
      "fast",
      "catalog",
      "typo-tolerance"
    ],
    "rank": 282,
    "elo": 1218,
    "eloCi": "±21",
    "winRate": 76,
    "starsDisplay": "59.5k",
    "weeklyStars": 13091,
    "weeklyGrowth": "+24.5%",
    "monthlyGrowth": "+12.4%",
    "wGrowthNum": 2.9,
    "mGrowthNum": 12.4,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-96%",
    "tokensNum": 96,
    "logoUrl": "https://github.com/meilisearch.png",
    "githubUrl": "https://github.com/meilisearch/meilisearch"
  },
  {
    "id": "manim",
    "name": "Manim",
    "repo": "3b1b/manim",
    "author": "3b1b",
    "persona": "Researchers & STEM",
    "targetRole": "STEM Educators & Visualizers",
    "stars": 94694,
    "wGrowth": 38.4,
    "mGrowth": 12.5,
    "website": "https://www.manim.community",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "150,000/mo",
    "searchVolNum": 150000,
    "searchTrend": "+22%",
    "latency": 80,
    "tokens": 88,
    "desc": "Grant Sanderson's (3Blue1Brown) programmatic mathematical animation engine for creating precise explanatory animations in Python.",
    "useCase": "Turn abstract calculus formulas, linear algebra transformations, and Fourier transforms into crisp SVG and MP4 animations.",
    "arch": "Python, Cairo, FFmpeg, LaTeX, ModernGL.",
    "tags": [
      "math-animation",
      "3blue1brown",
      "calculus",
      "visual-stem"
    ],
    "rank": 283,
    "elo": 1259,
    "eloCi": "±22",
    "winRate": 82,
    "starsDisplay": "94.7k",
    "weeklyStars": 28277,
    "weeklyGrowth": "+38.4%",
    "monthlyGrowth": "+12.5%",
    "wGrowthNum": 2.9,
    "mGrowthNum": 12.5,
    "latencyMs": "80ms",
    "latencyNum": 80,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/3b1b.png",
    "githubUrl": "https://github.com/3b1b/manim"
  },
  {
    "id": "appwrite",
    "name": "Appwrite",
    "repo": "appwrite/appwrite",
    "author": "appwrite",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Mobile & Web Founders",
    "stars": 57588,
    "wGrowth": 27.4,
    "mGrowth": 11.5,
    "website": "https://appwrite.io",
    "traffic": "1.6M/mo",
    "trafficNum": 1600000,
    "searchVol": "140,000/mo",
    "searchVolNum": 140000,
    "searchTrend": "+18%",
    "latency": 32,
    "tokens": 91,
    "desc": "Complete backend solution for web, mobile, and flutter developers. Database, Auth, Storage, Functions, and Messaging.",
    "useCase": "Deploy privacy-first user management, S3-compatible file storage, and serverless background workers on single VPS nodes.",
    "arch": "Docker, MariaDB, Redis, Swoole PHP runtime, Envoy Proxy.",
    "tags": [
      "self-hosted",
      "baas",
      "mobile-backend",
      "open-source"
    ],
    "rank": 284,
    "elo": 1212,
    "eloCi": "±23",
    "winRate": 75,
    "starsDisplay": "57.6k",
    "weeklyStars": 13653,
    "weeklyGrowth": "+27.4%",
    "monthlyGrowth": "+11.5%",
    "wGrowthNum": 2.8,
    "mGrowthNum": 11.5,
    "latencyMs": "32ms",
    "latencyNum": 32,
    "tokensSaved": "-91%",
    "tokensNum": 91,
    "logoUrl": "https://github.com/appwrite.png",
    "githubUrl": "https://github.com/appwrite/appwrite"
  },
  {
    "id": "bark-tts",
    "name": "Bark",
    "repo": "suno-ai/bark",
    "author": "suno-ai",
    "persona": "Music & Audio Creators",
    "targetRole": "Generative Audio Artists",
    "stars": 39272,
    "wGrowth": 9.7,
    "mGrowth": 12,
    "website": "https://github.com/suno-ai/bark",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "120,000/mo",
    "searchVolNum": 120000,
    "searchTrend": "+22%",
    "latency": 150,
    "tokens": 88,
    "desc": "Transformer-based text-to-audio model created by Suno. Generates speech, music, and ambient acoustics.",
    "useCase": "Generate expressive voice samples with laughs, sighs, and musical humming.",
    "arch": "EnCodec, GPT-style autoregressive transformer, PyTorch.",
    "tags": [
      "suno",
      "audio-generation",
      "speech-effects"
    ],
    "rank": 285,
    "elo": 1193,
    "eloCi": "±24",
    "winRate": 72,
    "starsDisplay": "39.3k",
    "weeklyStars": 4474,
    "weeklyGrowth": "+9.7%",
    "monthlyGrowth": "+12.0%",
    "wGrowthNum": 2.8,
    "mGrowthNum": 12,
    "latencyMs": "150ms",
    "latencyNum": 150,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/suno-ai.png",
    "githubUrl": "https://github.com/suno-ai/bark"
  },
  {
    "id": "directus",
    "name": "Directus",
    "repo": "directus/directus",
    "author": "directus",
    "persona": "Founders & Indie Hackers",
    "targetRole": "Backend Engineers & Founders",
    "stars": 38150,
    "wGrowth": 28,
    "mGrowth": 12,
    "website": "https://directus.io",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "95,000/mo",
    "searchVolNum": 95000,
    "searchTrend": "+15%",
    "latency": 45,
    "tokens": 90,
    "desc": "Open-source real-time API and app dashboard for managing any SQL database.",
    "useCase": "Provide instant REST and GraphQL APIs for client dashboards.",
    "arch": "Node.js, Vue.js, Knex.js, SQL database.",
    "tags": [
      "sql",
      "headless-api",
      "admin-panel"
    ],
    "rank": 286,
    "elo": 1181,
    "eloCi": "±25",
    "winRate": 70,
    "starsDisplay": "38.1k",
    "weeklyStars": 9184,
    "weeklyGrowth": "+28%",
    "monthlyGrowth": "+12.0%",
    "wGrowthNum": 2.8,
    "mGrowthNum": 12,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/directus.png",
    "githubUrl": "https://github.com/directus/directus"
  },
  {
    "id": "librosa",
    "name": "Librosa",
    "repo": "librosa/librosa",
    "author": "librosa",
    "persona": "Music & Audio Creators",
    "targetRole": "Audio Data Scientists",
    "stars": 8657,
    "wGrowth": 20.2,
    "mGrowth": 12,
    "website": "https://librosa.org",
    "traffic": "480k/mo",
    "trafficNum": 480000,
    "searchVol": "75,000/mo",
    "searchVolNum": 75000,
    "searchTrend": "+18%",
    "latency": 22,
    "tokens": 97,
    "desc": "Python package for music and audio analysis, providing building blocks for MIR (Music Information Retrieval).",
    "useCase": "Extract beat tempos, chroma features, harmonic content, and spectrograms from music tracks.",
    "arch": "NumPy, SciPy, Numba, SoundFile.",
    "tags": [
      "audio-analysis",
      "mir",
      "spectrogram",
      "music-science"
    ],
    "rank": 287,
    "elo": 1136,
    "eloCi": "±26",
    "winRate": 64,
    "starsDisplay": "8.7k",
    "weeklyStars": 1658,
    "weeklyGrowth": "+20.2%",
    "monthlyGrowth": "+12.0%",
    "wGrowthNum": 2.8,
    "mGrowthNum": 12,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/librosa.png",
    "githubUrl": "https://github.com/librosa/librosa"
  },
  {
    "id": "jax",
    "name": "JAX",
    "repo": "google/jax",
    "author": "google",
    "persona": "Researchers & STEM",
    "targetRole": "Machine Learning Researchers",
    "stars": 36380,
    "wGrowth": 22.1,
    "mGrowth": 10.8,
    "website": "https://github.com/google/jax",
    "traffic": "1.6M/mo",
    "trafficNum": 1600000,
    "searchVol": "120,000/mo",
    "searchVolNum": 120000,
    "searchTrend": "+18%",
    "latency": 18,
    "tokens": 98,
    "desc": "Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU.",
    "useCase": "Accelerate custom neural network gradient updates with XLA compilation.",
    "arch": "C++, Python, XLA compiler.",
    "tags": [
      "autograd",
      "tpu",
      "gpu",
      "linear-algebra"
    ],
    "rank": 288,
    "elo": 1179,
    "eloCi": "±18",
    "winRate": 70,
    "starsDisplay": "36.4k",
    "weeklyStars": 7325,
    "weeklyGrowth": "+22.1%",
    "monthlyGrowth": "+10.8%",
    "wGrowthNum": 2.5,
    "mGrowthNum": 10.8,
    "latencyMs": "18ms",
    "latencyNum": 18,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/google.png",
    "githubUrl": "https://github.com/google/jax"
  },
  {
    "id": "next-seo",
    "name": "Next SEO",
    "repo": "garmeeh/next-seo",
    "author": "garmeeh",
    "persona": "Marketers & Growth",
    "targetRole": "Next.js SEO Engineers",
    "stars": 8519,
    "wGrowth": 39.7,
    "mGrowth": 10.8,
    "website": "https://github.com/garmeeh/next-seo",
    "traffic": "450k/mo",
    "trafficNum": 450000,
    "searchVol": "35,000/mo",
    "searchVolNum": 35000,
    "searchTrend": "+16%",
    "latency": 24,
    "tokens": 97,
    "desc": "Next SEO is a plug in that makes managing your SEO easier in Next.js projects.",
    "useCase": "Inject OpenGraph, Twitter Cards, and Schema JSON-LD with zero boilerplate across routes.",
    "arch": "TypeScript, React, Next.js Head.",
    "tags": [
      "seo-schema",
      "opengraph",
      "structured-data"
    ],
    "rank": 289,
    "elo": 1132,
    "eloCi": "±19",
    "winRate": 63,
    "starsDisplay": "8.5k",
    "weeklyStars": 2571,
    "weeklyGrowth": "+39.7%",
    "monthlyGrowth": "+10.8%",
    "wGrowthNum": 2.5,
    "mGrowthNum": 10.8,
    "latencyMs": "24ms",
    "latencyNum": 24,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/garmeeh.png",
    "githubUrl": "https://github.com/garmeeh/next-seo"
  },
  {
    "id": "miniflux",
    "name": "Miniflux Reader",
    "repo": "miniflux/v2",
    "author": "miniflux",
    "persona": "Marketers & Growth",
    "targetRole": "Intelligence Operators",
    "stars": 9773,
    "wGrowth": 57.6,
    "mGrowth": 9.8,
    "website": "https://miniflux.app",
    "traffic": "320k/mo",
    "trafficNum": 320000,
    "searchVol": "15,000/mo",
    "searchVolNum": 15000,
    "searchTrend": "+14%",
    "latency": 22,
    "tokens": 98,
    "desc": "Miniflux is a minimalist and opinionated feed reader written in Go and PostgreSQL.",
    "useCase": "Consume 500+ competitor blogs and industry feeds every morning via clean JSON API.",
    "arch": "Go, PostgreSQL, micro-binary.",
    "tags": [
      "rss",
      "intelligence",
      "feed-reader"
    ],
    "rank": 290,
    "elo": 1130,
    "eloCi": "±20",
    "winRate": 63,
    "starsDisplay": "9.8k",
    "weeklyStars": 3715,
    "weeklyGrowth": "+57.6%",
    "monthlyGrowth": "+9.8%",
    "wGrowthNum": 2.3,
    "mGrowthNum": 9.8,
    "latencyMs": "22ms",
    "latencyNum": 22,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/miniflux.png",
    "githubUrl": "https://github.com/miniflux/v2"
  },
  {
    "id": "pyright",
    "name": "Pyright",
    "repo": "microsoft/pyright",
    "author": "microsoft",
    "persona": "Autonomous Software Engineers",
    "targetRole": "Python Architects",
    "stars": 15679,
    "wGrowth": 10.4,
    "mGrowth": 9.5,
    "website": "https://github.com/microsoft/pyright",
    "traffic": "1.1M/mo",
    "trafficNum": 1100000,
    "searchVol": "92,000/mo",
    "searchVolNum": 92000,
    "searchTrend": "+14%",
    "latency": 20,
    "tokens": 98,
    "desc": "Fast type checker for Python designed for large codebases and agentic tooling.",
    "useCase": "Verify agent-generated Python code types statically before test runs.",
    "arch": "TypeScript, Node.js, static type inference.",
    "tags": [
      "python-types",
      "static-analysis",
      "type-checker"
    ],
    "rank": 291,
    "elo": 1145,
    "eloCi": "±21",
    "winRate": 65,
    "starsDisplay": "15.7k",
    "weeklyStars": 1791,
    "weeklyGrowth": "+10.4%",
    "monthlyGrowth": "+9.5%",
    "wGrowthNum": 2.2,
    "mGrowthNum": 9.5,
    "latencyMs": "20ms",
    "latencyNum": 20,
    "tokensSaved": "-98%",
    "tokensNum": 98,
    "logoUrl": "https://github.com/microsoft.png",
    "githubUrl": "https://github.com/microsoft/pyright"
  },
  {
    "id": "transformers",
    "name": "Hugging Face Transformers",
    "repo": "huggingface/transformers",
    "author": "huggingface",
    "persona": "Researchers & STEM",
    "targetRole": "Machine Learning Researchers",
    "stars": 167014,
    "wGrowth": 21,
    "mGrowth": 9,
    "website": "https://huggingface.co",
    "traffic": "18.5M/mo",
    "trafficNum": 18500000,
    "searchVol": "1,200,000/mo",
    "searchVolNum": 1200000,
    "searchTrend": "+30%",
    "latency": 45,
    "tokens": 90,
    "desc": "State-of-the-art Machine Learning library providing thousands of pretrained models for PyTorch, TensorFlow, and JAX.",
    "useCase": "Fine-tune domain-specific BERT, RoBERTa, and Whisper checkpoints on proprietary scientific literature and lab corpora.",
    "arch": "Python, PyTorch, JAX, Safetensors, Tokenizers Rust core, ONNX.",
    "tags": [
      "deep-learning",
      "pytorch",
      "model-zoo",
      "nlp",
      "stem-research"
    ],
    "rank": 292,
    "elo": 1365,
    "eloCi": "±22",
    "winRate": 92,
    "starsDisplay": "167.0k",
    "weeklyStars": 31912,
    "weeklyGrowth": "+21%",
    "monthlyGrowth": "+9.0%",
    "wGrowthNum": 2.1,
    "mGrowthNum": 9,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/huggingface.png",
    "githubUrl": "https://github.com/huggingface/transformers"
  },
  {
    "id": "fastapi",
    "name": "FastAPI",
    "repo": "tiangolo/fastapi",
    "author": "tiangolo",
    "persona": "AI Ops & Tooling",
    "targetRole": "Python API Engineers",
    "stars": 102851,
    "wGrowth": 24.8,
    "mGrowth": 9.1,
    "website": "https://fastapi.tiangolo.com",
    "traffic": "5.2M/mo",
    "trafficNum": 5200000,
    "searchVol": "480,000/mo",
    "searchVolNum": 480000,
    "searchTrend": "+14%",
    "latency": 15,
    "tokens": 99,
    "desc": "FastAPI framework, high performance, easy to learn, fast to code, ready for production.",
    "useCase": "Serve agentic endpoints with automatic OpenAPI documentation and async execution.",
    "arch": "Starlette, Pydantic, Python async.",
    "tags": [
      "python",
      "api",
      "openapi",
      "async"
    ],
    "rank": 293,
    "elo": 1280,
    "eloCi": "±23",
    "winRate": 85,
    "starsDisplay": "102.9k",
    "weeklyStars": 22181,
    "weeklyGrowth": "+24.8%",
    "monthlyGrowth": "+9.1%",
    "wGrowthNum": 2.1,
    "mGrowthNum": 9.1,
    "latencyMs": "15ms",
    "latencyNum": 15,
    "tokensSaved": "-99%",
    "tokensNum": 99,
    "logoUrl": "https://github.com/tiangolo.png",
    "githubUrl": "https://github.com/tiangolo/fastapi"
  },
  {
    "id": "docsearch",
    "name": "Algolia DocSearch",
    "repo": "algolia/docsearch",
    "author": "algolia",
    "persona": "Marketers & Growth",
    "targetRole": "DevRel & Content Leads",
    "stars": 4377,
    "wGrowth": 2.1,
    "mGrowth": 9,
    "website": "https://docsearch.algolia.com",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "28,000/mo",
    "searchVolNum": 28000,
    "searchTrend": "+14%",
    "latency": 30,
    "tokens": 95,
    "desc": "The easiest way to add search to your developer documentation. Free for technical documentation.",
    "useCase": "Index entire open-source documentation trees into instantaneous search modals.",
    "arch": "Scrapy scraper, Algolia search API, React modal.",
    "tags": [
      "documentation-search",
      "devrel",
      "crawling"
    ],
    "rank": 294,
    "elo": 1128,
    "eloCi": "±24",
    "winRate": 63,
    "starsDisplay": "4.4k",
    "weeklyStars": 130,
    "weeklyGrowth": "+2.1%",
    "monthlyGrowth": "+9.0%",
    "wGrowthNum": 2.1,
    "mGrowthNum": 9,
    "latencyMs": "30ms",
    "latencyNum": 30,
    "tokensSaved": "-95%",
    "tokensNum": 95,
    "logoUrl": "https://github.com/algolia.png",
    "githubUrl": "https://github.com/algolia/docsearch"
  },
  {
    "id": "ghost-cms",
    "name": "Ghost CMS",
    "repo": "TryGhost/Ghost",
    "author": "TryGhost",
    "persona": "Marketers & Growth",
    "targetRole": "Content & SEO Directors",
    "stars": 55495,
    "wGrowth": 17.6,
    "mGrowth": 8.5,
    "website": "https://ghost.org",
    "traffic": "4.2M/mo",
    "trafficNum": 4200000,
    "searchVol": "240,000/mo",
    "searchVolNum": 240000,
    "searchTrend": "+12%",
    "latency": 40,
    "tokens": 92,
    "desc": "Modern open-source publishing platform designed for modern online media, content marketing, and email newsletters.",
    "useCase": "Dominate organic Google rankings with clean Schema markup, fast CDN caching, and integrated paid subscription paywalls.",
    "arch": "Node.js, Express, MySQL, Handlebars/Next.js frontend, Mailgun.",
    "tags": [
      "seo-blog",
      "newsletter",
      "publishing",
      "organic-traffic"
    ],
    "rank": 295,
    "elo": 1210,
    "eloCi": "±25",
    "winRate": 75,
    "starsDisplay": "55.5k",
    "weeklyStars": 9239,
    "weeklyGrowth": "+17.6%",
    "monthlyGrowth": "+8.5%",
    "wGrowthNum": 2,
    "mGrowthNum": 8.5,
    "latencyMs": "40ms",
    "latencyNum": 40,
    "tokensSaved": "-92%",
    "tokensNum": 92,
    "logoUrl": "https://github.com/TryGhost.png",
    "githubUrl": "https://github.com/TryGhost/Ghost"
  },
  {
    "id": "sympy",
    "name": "SymPy",
    "repo": "sympy/sympy",
    "author": "sympy",
    "persona": "Researchers & STEM",
    "targetRole": "Physicists & Engineers",
    "stars": 14994,
    "wGrowth": 17.1,
    "mGrowth": 8,
    "website": "https://www.sympy.org",
    "traffic": "1.2M/mo",
    "trafficNum": 1200000,
    "searchVol": "180,000/mo",
    "searchVolNum": 180000,
    "searchTrend": "+11%",
    "latency": 25,
    "tokens": 97,
    "desc": "A Python library for symbolic mathematics, algebra, calculus, and matrices.",
    "useCase": "Compute exact analytical derivatives and solve differential equations.",
    "arch": "Pure Python, symbolic expression tree.",
    "tags": [
      "symbolic-math",
      "algebra",
      "calculus"
    ],
    "rank": 296,
    "elo": 1140,
    "eloCi": "±26",
    "winRate": 65,
    "starsDisplay": "15.0k",
    "weeklyStars": 2437,
    "weeklyGrowth": "+17.1%",
    "monthlyGrowth": "+8.0%",
    "wGrowthNum": 1.9,
    "mGrowthNum": 8,
    "latencyMs": "25ms",
    "latencyNum": 25,
    "tokensSaved": "-97%",
    "tokensNum": 97,
    "logoUrl": "https://github.com/sympy.png",
    "githubUrl": "https://github.com/sympy/sympy"
  },
  {
    "id": "stable-diffusion-webui",
    "name": "Stable Diffusion WebUI",
    "repo": "AUTOMATIC1111/stable-diffusion-webui",
    "author": "AUTOMATIC1111",
    "persona": "Music & Audio Creators",
    "targetRole": "Album Art & Visual Artists",
    "stars": 165204,
    "wGrowth": 16.3,
    "mGrowth": 7.5,
    "website": "https://github.com/AUTOMATIC1111/stable-diffusion-webui",
    "traffic": "6.5M/mo",
    "trafficNum": 650000,
    "searchVol": "680,000/mo",
    "searchVolNum": 680000,
    "searchTrend": "+15%",
    "latency": 120,
    "tokens": 88,
    "desc": "Browser interface based on Gradio library for Stable Diffusion text-to-image and inpainting generation.",
    "useCase": "Generate 4K album cover art, song poster backdrops, and automated social visual assets from lyric style prompts.",
    "arch": "Python, PyTorch, Gradio, TorchScript, xFormers, ControlNet pipeline.",
    "tags": [
      "generative-art",
      "album-covers",
      "image-generation",
      "gradio",
      "visuals"
    ],
    "rank": 297,
    "elo": 1365,
    "eloCi": "±18",
    "winRate": 92,
    "starsDisplay": "165.2k",
    "weeklyStars": 25760,
    "weeklyGrowth": "+16.3%",
    "monthlyGrowth": "+7.5%",
    "wGrowthNum": 1.8,
    "mGrowthNum": 7.5,
    "latencyMs": "120ms",
    "latencyNum": 120,
    "tokensSaved": "-88%",
    "tokensNum": 88,
    "logoUrl": "https://github.com/AUTOMATIC1111.png",
    "githubUrl": "https://github.com/AUTOMATIC1111/stable-diffusion-webui"
  },
  {
    "id": "strapi",
    "name": "Strapi",
    "repo": "strapi/strapi",
    "author": "strapi",
    "persona": "Marketers & Growth",
    "targetRole": "Content Strategists",
    "stars": 73288,
    "wGrowth": 13.6,
    "mGrowth": 7.5,
    "website": "https://strapi.io",
    "traffic": "3.2M/mo",
    "trafficNum": 3200000,
    "searchVol": "310,000/mo",
    "searchVolNum": 310000,
    "searchTrend": "+9%",
    "latency": 48,
    "tokens": 89,
    "desc": "The leading open-source headless CMS. 100% JavaScript / TypeScript and fully customizable.",
    "useCase": "Power content hubs and localized SEO blogs with structured REST endpoints.",
    "arch": "Node.js, React, Koa, SQL.",
    "tags": [
      "cms",
      "headless",
      "content-api"
    ],
    "rank": 298,
    "elo": 1243,
    "eloCi": "±19",
    "winRate": 79,
    "starsDisplay": "73.3k",
    "weeklyStars": 9949,
    "weeklyGrowth": "+13.6%",
    "monthlyGrowth": "+7.5%",
    "wGrowthNum": 1.8,
    "mGrowthNum": 7.5,
    "latencyMs": "48ms",
    "latencyNum": 48,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/strapi.png",
    "githubUrl": "https://github.com/strapi/strapi"
  },
  {
    "id": "spleeter",
    "name": "Spleeter",
    "repo": "deezer/spleeter",
    "author": "deezer",
    "persona": "Music & Audio Creators",
    "targetRole": "Remix Artists & DJs",
    "stars": 28477,
    "wGrowth": 14.8,
    "mGrowth": 7.8,
    "website": "https://github.com/deezer/spleeter",
    "traffic": "850k/mo",
    "trafficNum": 850000,
    "searchVol": "160,000/mo",
    "searchVolNum": 160000,
    "searchTrend": "+8%",
    "latency": 120,
    "tokens": 90,
    "desc": "Deezer source separation audio library trained on TensorFlow for 2, 4, and 5 stem extraction.",
    "useCase": "Quickly split MP3 tracks into vocals and accompaniment on CPU.",
    "arch": "TensorFlow, U-Net architecture, Librosa.",
    "tags": [
      "stem-splitter",
      "deezer",
      "acapella-extraction"
    ],
    "rank": 299,
    "elo": 1163,
    "eloCi": "±20",
    "winRate": 68,
    "starsDisplay": "28.5k",
    "weeklyStars": 4123,
    "weeklyGrowth": "+14.8%",
    "monthlyGrowth": "+7.8%",
    "wGrowthNum": 1.8,
    "mGrowthNum": 7.8,
    "latencyMs": "120ms",
    "latencyNum": 120,
    "tokensSaved": "-90%",
    "tokensNum": 90,
    "logoUrl": "https://github.com/deezer.png",
    "githubUrl": "https://github.com/deezer/spleeter"
  },
  {
    "id": "matomo",
    "name": "Matomo Analytics",
    "repo": "matomo-org/matomo",
    "author": "matomo-org",
    "persona": "Marketers & Growth",
    "targetRole": "Enterprise Marketing Leads",
    "stars": 21930,
    "wGrowth": 10.8,
    "mGrowth": 7.7,
    "website": "https://matomo.org",
    "traffic": "1.4M/mo",
    "trafficNum": 1400000,
    "searchVol": "140,000/mo",
    "searchVolNum": 140000,
    "searchTrend": "+9%",
    "latency": 45,
    "tokens": 89,
    "desc": "The Google Analytics alternative that protects your data and user privacy.",
    "useCase": "Self-host full visitor tracking, heatmaps, and funnel analytics on your own cloud.",
    "arch": "PHP, MySQL, Vue.js.",
    "tags": [
      "google-analytics-alt",
      "enterprise",
      "heatmaps"
    ],
    "rank": 300,
    "elo": 1153,
    "eloCi": "±21",
    "winRate": 66,
    "starsDisplay": "21.9k",
    "weeklyStars": 2486,
    "weeklyGrowth": "+10.8%",
    "monthlyGrowth": "+7.7%",
    "wGrowthNum": 1.8,
    "mGrowthNum": 7.7,
    "latencyMs": "45ms",
    "latencyNum": 45,
    "tokensSaved": "-89%",
    "tokensNum": 89,
    "logoUrl": "https://github.com/matomo-org.png",
    "githubUrl": "https://github.com/matomo-org/matomo"
  }
]