colibri
colibri is 03 colibri → runs a 744B model on 25GB of RAM by reading it off your disk. It is ranked #403 on the AI Radar, in Open & local models, first seen 52 days ago and shared in 3 posts (773k views).
What colibri says about itself
Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦 - JustVugg/colibri
GitHub - JustVugg/colibri: Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦 · GitHub Skip to content Navigation Menu Sign in Appearance settings Search / Sign in Sign up Appearance settings You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window. Reload to refresh your session. Dismiss alert {{ message }} Repository navigation colibri Public Notifications You must be signed in to change notification settings Fork 4.9k (…
What people said about colibri on X
8 free GitHub repos that take your AI off Big Tech's servers. Each one replaces an app that keeps your chats, photos or camera footage on a company's computers. 792,000 GitHub stars between them. 01 Ollama → runs DeepSeek, Qwen, Gemma and gpt-oss on your own laptop: http://github.com/ollama/ollama 02 Open WebUI → the…
— @alex_verem, 2 days ago · 239 likes · see the post
Someone built a free, open-source inference engine that lets you run frontier-scale mixture-of-experts models on a normal PC, so you can use big models without a datacenter. → Keeps the most-used experts on the GPU, and the rest in RAM or on disk → The README lists 8 GB of RAM as the minimum → Apache-2.0, 47K+ stars,…
— @RoundtableSpace, 2 days ago · 52 likes · see the post
Alternatives to colibri
- kimi-k3.1-pro — Releasing the model weights of Kimi K3.1 Pro
- unsloth.ai — Guide to running Mistral Large 4 GGUFs locally with Unsloth and llama.cpp.
- niklaslenz.com — Play the world in your browser. Every number, the exact command and the raw output
- qwen3.8-flash-next-gguf — Image-Text-to-Text model with 177B parameters. Download, explore files and run it on Hugging Face.
- project-maya — Plus a round of fixes, UI improvements and community PRs.
- twil-8b — Text Generation model with 9B parameters. Download, explore files and run it on Hugging Face.
colibri in numbers
- Rank on the AI Radar: #403 of 10410
- Shared in 3 posts by 3 accounts: @alex_verem, @RoundtableSpace
- 773k views on those posts
- First seen 52 days ago, last shared 2 days ago
- Pricing seen by AGK Intelligence: open source
- Market: Open & local models
FAQ
What is colibri?
03 colibri → runs a 744B model on 25GB of RAM by reading it off your disk It was first shared on X 52 days ago and is ranked #403 on the AI Radar.
Is colibri free?
It is open source.
Who shared colibri?
3 accounts on X, including @alex_verem, @RoundtableSpace, in 3 posts totalling 773k views.
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 65.6k posts from 5k X accounts over the last 14 days, 10.4k tools, 19 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-06 16:05 UTC. Full method.