Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Compare compatible open-source local LLM projects by their metrics to assess popularity and activeness.
| Date | Stars |
|---|---|
| 2026-07-31 | 789 |
| 2026-08-03 | 790 |
| 2026-08-06 | 790 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# 👋 Awesome Local LLMs A curated list of open-source projects for building with LLMs — spanning the whole stack, from model weights and training code to inference runtimes, agent frameworks, end-user apps, and the lists and tutorials that help you learn. GitHub metrics (stars, forks, issues, contributors, releases, time since last commit) are refreshed automatically every week. Projects are organised into eight categories: - **AI Engineering** — agent SDKs, orchestration, prompting & scaffolding, memory/RAG, tools & integrations, MCP, evaluation, and guardrails - **Applications** — coding assistants, personal agents, productivity, content creation, research, data analytics, and automation - **Infrastructure** — local runtimes, production serving, distributed & edge inference, optimisation, gateways, observability, and vector search - **Model Development** — fine-tuning, training, RLHF, deep learning frameworks, dataset engineering, and interpretability - **Models** — foundation, vision, audio/speech, embedding, on-device, and domain-specific model releases - **Lists** — curated collections of prompts, tools, models, papers, and datasets - **Tutorials** — getting-started guides, courses, roadmaps, and interview prep - **Misc** — everything else Each repo is tagged with one **category**, one or more **subcategories**, and a set of cross-cutting **keywords** (techniques, integrations, modalities, and domains such as `RAG`, `MCP`, or `Context engineering`). **Contributions are welcome!** Suggest a repo I've missed by opening an issue. There is also a fuller table of metrics in this [Google Sheet](https://docs.google.com/spreadsheets/d/1Xv38p90V3GiJXjq0a3qc24056Vicn1I5MG6QiFE6nVE/edit?usp=sharing) and [Airtable](https://airtable.com/apparaKqezkq2LECD/shrE26kWFaVU1cvgb) _(no longer updated — kept for reference)_. For my thoughts on local LLM tooling: <https://vinlam.com/posts/local-llm-options/> Note the condensed table below shows the **top 100 projects by stars**, filtered to repos with a commit within the last 60 days. ## Open-Source LLM & Agent Projects *Last Updated: 27/07/2026* <!-- BEGIN_TABLE --> | # | Repo | Category | Tags | About | Stars | Forks | Issues | Contributors | Releases | License | Time Since Last Commit | |----------|----------|----------|----------|----------|----------|----------|----------|----------|----------|----------|----------| | 1 | [openclaw](https://github.com/openclaw/openclaw) | Applications | Coding Agents & Assistants | Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞 | 384,293 | 80,738 | 4,034 | 0 | 226 | Other | 0 days, 0 hrs, 1 mins | | 2 | [superpowers](https://github.com/obra/superpowers) | AI Engineering | Tools & Integrations | An agentic skills framework & software development methodology that works. | 261,841 | 23,380 | 158 | 0 | 11 | MIT License | 2 days, 12 hrs, 41 mins | | 3 | [ECC](https://github.com/affaan-m/ECC) | AI Engineering | Tools & Integrations | The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. | 233,833 | 35,652 | 21 | 0 | 14 | MIT License | 0 days, 6 hrs, 17 mins | | 4 | [hermes-agent](https://github.com/NousResearch/hermes-agent) | AI Engineering | Agent SDK / Framework, Agent Frontends / Chat UI | The agent that grows with you | 221,189 | 42,203 | 8,190 | 0 | 22 | MIT License | 0 days, 3 hrs, 49 mins | | 5 | [n8n](https://github.com/n8n-io/n8n) | AI Engineering | Visual / Low-Code Builder, Multi-Agent Orchestration | Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations. | 198,185 | 59,656 | 324 | 0 | 737 | Other | 0 days, 0 hrs, 0 mins | | 6 | [skills](https://github.com/mattpocock/skills) | AI Engineering | Tools & Integrations | Skills for Real Eng
Excerpt of 35,605 characters
Read on GitHub
125
14
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:d23dae408099d732, desc:local llm