colibri
colibri runs large open mixture-of-experts language models locally by streaming experts from disk, so the whole model need not fit in RAM. Use its dashboard or compatible APIs.
Pick a small issue you can reproduce, then fix it with your AI coding agent.
Share on XLicense: Apache-2.0
Overview
colibri is a pure-C local inference engine for running large open mixture-of-experts language models on computers people already own. It keeps the resident model portion and an expert cache in RAM, while streaming other experts from disk, reducing the memory needed to use models too large to load at once. Setup checks the hardware, recommends a fitting model, and prepares the engine and model. Use the browser dashboard or connect apps through OpenAI- and Anthropic-compatible APIs.
Key features
- Checks hardware and recommends a model that fits available RAM and disk space.
- Streams experts from disk as needed instead of loading the whole model into RAM.
- Builds an engine locally when a compiler is available, or downloads a prebuilt engine.
- Downloads models with progress and resume, and provides a browser dashboard plus OpenAI- and Anthropic-compatible API endpoints.
Best for
People who want to run large open language models on a computer they already own, including systems without a GPU. The computer needs enough RAM and disk space for the selected model.
- Upstream
- JustVugg/colibri
- Fork on GitHub
- Guo-astro/colibri
- Upstream stars
- 48k
- Category
- AI agents and LLM tools
- Language
- C
- License
- Apache-2.0
- Forked
- 2026-10-12
- Sync status
- In syncLast synced 2026-11-08
More in AI agents and LLM tools
/last30days is an AI-agent skill for researching recent discussions across social platforms, video, news, markets, and the web. It brings scattered signals together in a synthesized brief for trend research and preparation.
Forked 2026-10-22Last synced 2026-11-08License: MITAI agents and LLM toolsGitHub
A portable C99 inference engine for running Kimi K3 on a CPU without a GPU or machine-learning framework. It streams the large checkpoint from disk to support inference with limited RAM, for batch text generation or text chat.
Forked 2026-10-21Last synced 2026-11-08License: Apache-2.0AI agents and LLM toolsGitHub
LayerFS gives AI agents writable workspaces over shared history, so they can work in parallel without copying the base. Use it to checkpoint, branch, rewind, and merge agent changes.
Forked 2026-10-21Last synced 2026-11-08License: MITAI agents and LLM toolsGitHub
Magnitude runs open models locally and tunes kernels for your hardware. Use it to connect local models to supported agents or an OpenAI-compatible client.
Forked 2026-10-19Last synced 2026-11-08License: Apache-2.0AI agents and LLM toolsGitHub