Nehanth/pooled
Peer-to-peer browser-based LLM inference that pools multiple devices to run large open models.
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history
Pooled runs a single open model across the devices in a room, with each device holding some of the model’s layers and contributing compute via WebGPU and WebRTC. It supports chat interactions and a Code mode for building and fixing small apps directly in the browser tab. The system uses speculative decoding and WGSL shaders, requires no installation or account, and performs all inference locally across the pooled peers.
Frequently asked
- What is Nehanth/pooled?
- Peer-to-peer browser-based LLM inference that pools multiple devices to run large open models.
- Is pooled open source?
- Yes — Nehanth/pooled is open source, released under the MIT license.
- What language is pooled written in?
- Nehanth/pooled is primarily written in JavaScript.
- How popular is pooled?
- Nehanth/pooled has 538 stars on GitHub.
- Where can I find pooled?
- Nehanth/pooled is on GitHub at https://github.com/Nehanth/pooled.