AdvertiseWatchSubscribe
Updated , rescanned every two hoursMonday, November 2, 2026

▲ 655 today46.2k starsCHN ×1

Open-source C engine streams massive MoE models from your disk

LLMsInference

JustVugg/colibri: preview

What is colibri?

colibri lets you run huge mixture-of-experts language models on a regular PC by keeping only the active part of the model in RAM and loading the rest from storage on demand. It’s a single-file C runtime that works on CPU, Vulkan or CUDA without extra dependencies.

Who it is for

Developers and hobbyists who want to experiment with state-of-the-art LLMs without buying expensive GPUs can spin up these models locally.

Get the signal first.

Join the list for the morning email: the day's biggest AI launches, explained in five minutes. Unsubscribe any time.

Prefer a reader? Follow the RSS feed.