Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
README.md | 2025-07-24 | 654 Bytes | |
Release v0.8.7 source code.tar.gz | 2025-07-24 | 368.3 kB | |
Release v0.8.7 source code.zip | 2025-07-24 | 452.3 kB | |
Totals: 3 Items | 821.2 kB | 1 |
What's Changed
- Support adustable min_lr from DPO training script by @viswavi in https://github.com/OpenRLHF/OpenRLHF/pull/1092
- Support soft overlong punishment from DAPO by @xjli360 in https://github.com/OpenRLHF/OpenRLHF/pull/1091
- Support token in token out and agent loop by @xiaoxigua999 @physicsru in https://github.com/OpenRLHF/OpenRLHF/pull/1094
New Contributors
- @viswavi made their first contribution in https://github.com/OpenRLHF/OpenRLHF/pull/1092
- @xjli360 made their first contribution in https://github.com/OpenRLHF/OpenRLHF/pull/1091
Full Changelog: https://github.com/OpenRLHF/OpenRLHF/compare/v0.8.6...v0.8.7