Pool spare GPU capacity to run LLMs at larger scale(github.com)

11 ポイント投稿者 i3864 か月前3 コメント

https://github.com/michaelneale/mesh-llm

3 コメント

lostmsu4 か月前

> MoE models via expert sharding with zero cross-node inference traffic

This makes the whole project questionable

vagrantJin4 か月前

This is very promising, definitely looks more user friendly than exo. Can't wait to try it out.

iwinux4 か月前

You lost me on "spare GPU". I don't have any capable GPUs, let alone spare ones :)