[Growth] AI/LLM Collection Expansion - Track 150+ Top AI Repos with Growth Metrics

[sykp241095]

🎯 Opportunity

The AI/LLM ecosystem is exploding with 150+ high-growth repositories that OSSInsight should track systematically.

Market Analysis (March 2026)

Top AI Repos by Stars:

• ollama/ollama: 165,913 ⭐ (Go) - Local LLM inference
• huggingface/transformers: 158,275 ⭐ (Python) - ML framework
• prompts.chat: 153,909 ⭐ (HTML) - AI prompt playground
• langgenius/dify: 134,027 ⭐ (TypeScript) - LLM app platform
• langchain-ai/langchain: 130,671 ⭐ (Python) - Agent framework
• open-webui/open-webui: 128,326 ⭐ (Python) - LLM UI

Agent Framework Leaders:

• FoundationAgents/MetaGPT: 65,839 ⭐ - Multi-agent collaboration
• pydantic/pydantic-ai: 15,681 ⭐ - Type-safe AI agents
• TransformerOptimus/SuperAGI: 17,293 ⭐ - AGI toolkit
• camel-ai/camel: 16,458 ⭐ - Multi-agent framework
• QwenLM/Qwen-Agent: 15,735 ⭐ - Qwen agent framework

MCP Server Ecosystem (Fastest Growing):

• punkpeye/awesome-mcp-servers: 83,861 ⭐ - MCP registry
• microsoft/playwright-mcp: 29,478 ⭐ - Browser automation
• github/github-mcp-server: 28,161 ⭐ - GitHub integration
• PrefectHQ/fastmcp: 23,902 ⭐ - Python MCP framework
• oraios/serena: 21,939 ⭐ - Coding toolkit

RAG & Vector DBs:

• infiniflow/ragflow: 75,843 ⭐ - RAG engine
• run-llama/llama_index: 47,885 ⭐ - RAG framework
• milvus-io/milvus: 43,461 ⭐ - Vector database
• zilliztech/gpt-cache: 12,500 ⭐ - LLM caching

Gap Analysis

Current AI collection is incomplete. Missing systematic tracking of:

1. Inference engines (llama.cpp, vllm, gpt4all, ollama)
2. RAG platforms (ragflow, firecrawl, llama_index)
3. Agent monitoring (agentops, langfuse)
4. Mobile agents (AppAgent, lamda)
5. MCP servers (50+ new servers in 2025-2026)
6. AI gateways (litellm, portkey, ai-gateway)

📋 Tasks

• Expand AI/LLM collection to 150+ repos across 8 categories:
  • Inference Engines (15 repos)
  • Agent Frameworks (25 repos)
  • RAG Platforms (20 repos)
  • MCP Servers (40 repos)
  • Vector Databases (15 repos)
  • AI Gateways (15 repos)
  • Monitoring & Observability (10 repos)
  • Developer Tools (10 repos)
• Add growth velocity tracking (stars/week, PR velocity, contributor growth)
• Create category leaderboards with filterable views
• Build "AI Ecosystem Map" visualization showing relationships
• Add comparison views (head-to-head metrics for competing projects)
• Create "AI Trending Weekly" automated report

🚀 Growth Impact

• Capture AI developer traffic (largest growing segment in open source)
• Enable AI project maintainers to track competitors and benchmark
• Create viral "State of AI Open Source" quarterly reports
• Position OSSInsight as the definitive AI ecosystem analytics platform
• Drive 10K+ monthly views from AI developer community

📊 Success Metrics

• 150+ AI projects tracked with comprehensive growth metrics
• 10K+ monthly views on AI collection pages
• 5+ AI projects using OSSInsight badges in README
• 3+ AI newsletters featuring OSSInsight data
• 500+ AI developers creating accounts to track projects

🔗 References

• AI Collection: https://ossinsight.io/collections/ai
• MCP Registry: https://github.com/modelcontextprotocol/registry
• Hugging Face Trends: https://huggingface.co/trends

[sykp241095]

AI/LLM Collection Expansion Analysis (March 2026)

Comprehensive analysis completed. Here are the top projects to add:

Agent Frameworks (Priority: High)

• crewAIInc/crewAI (46.9k ⭐) - Multi-agent collaboration
• langchain-ai/langgraph (27.2k ⭐) - Graph-based orchestration
• FlowiseAI/Flowise (51k ⭐) - Visual agent builder
• conductor-oss/conductor (31.6k ⭐) - Event-driven agentic platform

RAG Systems (Priority: High)

• infiniflow/ragflow (75.8k ⭐) - Production RAG engine
• langgenius/dify (134k ⭐) - Agentic workflow platform
• microsoft/graphrag (31.7k ⭐) - Graph-based RAG
• HKUDS/LightRAG (30.2k ⭐) - Lightweight RAG

Vector Databases (Priority: Medium)

• milvus-io/milvus (43.5k ⭐) - Cloud-native vector DB
• qdrant/qdrant (29.8k ⭐) - Rust-based, massive scale
• weaviate/weaviate (15.9k ⭐) - Structured filtering + vector

Memory & Context (Priority: Medium)

• mem0ai/mem0 (50.8k ⭐) - Universal memory layer
• getzep/graphiti (24.1k ⭐) - Real-time knowledge graphs

Browser Automation (Priority: Medium)

• browser-use/browser-use (82.9k ⭐) - Web automation for agents
• vercel-labs/agent-browser (24.3k ⭐) - Rust-based CLI

Total identified: 150+ repos across 5 categories. Recommend organizing into themed sub-collections for better navigation.

---

Generated by OSSInsight Growth Analysis Bot

[sykp241095]

Done: 28 new AI/LLM collections created in PRs #2271/#2277.