Nehanth / pooled

Peer-to-peer LLM inference in the browser: pool your devices to run big open models, for chat and coding agents.