best-of ranking · updated nightly

Best Open Source AI & Machine Learning Tools

Ranked by GitHub stars — actual adoption, not editorial taste. All 10 are free, open source and self-hostable; license and last activity on every entry.

1. 1
   OpenClaw ★ 390K

   An AI assistant that acts, not just answers

   NOASSERTION TypeScript pushed 2026-09-17 details →
2. 2
   Hermes Agent ★ 246K

   Desktop AI agent with persistent memory across platforms

   MIT Python pushed 2026-09-17 details →
3. 3
   Ollama ★ 181K

   Run open-source LLMs locally on your own machine

   MIT Go pushed 2026-09-17 details →
4. 4
   Dify ★ 156K

   Build and deploy AI agents and RAG pipelines visually

   NOASSERTION TypeScript pushed 2026-09-17 details →
5. 5
   Open WebUI ★ 152K

   One interface for every AI model you run or connect

   NOASSERTION Python pushed 2026-09-17 details →
6. 6
   langchain ★ 147K

   The agent engineering platform.

   MIT Python pushed 2026-09-17 details →
7. 7
   ponytail ★ 141K

   Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.

   MIT JavaScript pushed 2026-09-14 details →
8. 8
   llama.cpp ★ 129K

   Run LLMs locally with minimal setup, maximum hardware support

   MIT C++ pushed 2026-09-17 details →
9. 9
   generative-ai-for-beginners ★ 120K

   21 Lessons, Get Started Building with Generative AI

   MIT Jupyter Notebook pushed 2026-09-17 details →
10. 10
    graphify ★ 119K

    Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini

    Apache-2.0 Python pushed 2026-09-16 details →

last updated nightly · ranking method: github stars, live-tracked

Looking for a replacement for one specific commercial product? Browse alternatives by tool →