A scalable asynchronous reinforcement learning implementation with in-flight weight updates.
active 2025-04-22 → 2026-07-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 455 days from 2025-04-22 to 2026-07-20. Pushes: 621 total, peak 18 in a day. Pull requests: 94 total, peak 5 in a day. Issues: 17 total, peak 3 in a day. Comments: 102 total, peak 7 in a day. Stars: 214 total, peak 39 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| AlexPiche | 270 | 235 | 23 | 8 |
| rizar | 193 | 99 | 31 | 37 |
| rafapi | 167 | 125 | 6 | 20 |
| ollmer | 54 | 24 | 7 | 13 |
| NicolasAG | 44 | 40 | 2 | 1 |
| ehsk | 41 | 14 | 7 | 10 |
| bigximik | 19 | 19 | 0 | 0 |
| jlamypoirier | 19 | 19 | 0 | 0 |
| Alex-Gurung | 10 | 10 | 0 | 0 |
| kashif | 8 | 0 | 2 | 4 |
| shivakrishnareddy-ma | 6 | 5 | 1 | 0 |
| shruthan | 4 | 2 | 2 | 0 |
| oleksost | 4 | 4 | 0 | 0 |
| Copilot | 3 | 0 | 0 | 1 |
| kushasareen | 2 | 0 | 1 | 1 |
| optimass | 2 | 0 | 0 | 0 |
| lewtun | 2 | 0 | 0 | 1 |
| RaymondLi0 | 1 | 1 | 0 | 0 |
| joanrod | 1 | 0 | 1 | 0 |
| Vattikondadheeraj | 1 | 0 | 1 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#134ehsk2026-04-15 16:37
- Pull request#131taivu19982026-04-02 19:03
- Pull request#130rafapi2026-03-06 19:56
- Issue#96rafapi2026-02-13 11:24Simplify Codebase with Native vLLM Weight Update Support
- Pull request#127ehsk2026-02-06 20:50
- Pull request#125ehsk2026-02-03 21:01
- Pull request#123ehsk2026-01-27 09:22
- Issue comment#121ehsk2026-01-18 18:52vLLM upgrade
- Pull request#121ehsk2025-12-23 14:36
- Pull request#120shruthan2025-12-22 18:23
- Issue comment#118kushasareen2025-12-20 05:50'utf-8' codec can't decode byte 0xe2 in position 0: unexpected end of data
- Pull request#119kushasareen2025-12-18 21:29
- Issue comment#116rafapi2025-12-18 12:00if eval instruction id list
- Pull request#117Vattikondadheeraj2025-12-16 22:03
- Pull request#116shruthan2025-12-16 18:56
- Pull request#114rafapi2025-12-15 18:29
- Issue comment#113lewtun2025-12-08 20:44Could not decode JSON from weight_update_request/0/0
- Issue#113lewtun2025-12-08 20:28Could not decode JSON from weight_update_request/0/0
- Pull request#112rafapi2025-12-05 17:25
- Issue comment#105rafapi2025-11-30 12:32Implement multi-environment and multi-domain mixing
- Issue comment#48rafapi2025-11-28 09:41Debug the Miniwob example
- Issue comment#48ollmer2025-11-27 18:29Debug the Miniwob example
- Pull request#108rafapi2025-11-23 18:26
- Issue comment#107rafapi2025-11-22 20:23GSPO
- Issue comment#103rafapi2025-11-18 17:52Deps cleanup
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 214 stars here means stars gained during the window, not the repo's star count.