Nehanth / pooled
Peer-to-peer LLM inference in the browser: pool your devices to run big open models, for chat and coding agents.
Why this job: says: llm inference; topics: local-llm, inference
What it does
Friends open one link. Each device (laptop, desktop, phone) holds some of the model's layers and runs them on its own GPU with our own WGSL kernels. For every token, a small hidden state (10 KB on the 27B, 4 KB on the MoE) passes from tab to tab over direct WebRTC connections, and the host turns the result into the next word.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 6 days ago
- Stars
- 637
- +13 this week
- Licence
- MIT
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v2.0.0
- Released 24 days ago
- Built with
- JavaScript
- 90% of the code
Is it looked after?
- Contributors
- 3
- Last 90 days
- Issues
- 65% closed
- 56 still open
- Pull requests
- 241 merged
- 10 waiting
- Releases
- 3
- 3 in the last year
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| JustVugg / lumabri | 216 | 3 days ago | Apache-2.0 | No downloads |
| mlc-ai / web-llm | 19,365 | 6 days ago | Apache-2.0 | No downloads |
| gpustack / gpustack | 6,004 | today | Apache-2.0 | No downloads |
More: languages, topics and project record
Languages
- JavaScript 90%
- MDX 4%
- HTML 4%
- CSS 1%
Topics
- browser
- distributed-inference
- inference
- llm
- local-llm
- peer-to-peer
- speculative-decoding
- webgpu
- webrtc
- wgsl
Project record
- JobLocal LLMs
- StatusListed
- Commits681
- Forks97
- Watching2
- Stars in 28 days+53
- CreatedSep 1, 2026
- First seen by usNov 7, 2026
- CheckedNov 7, 2026