{"slug":"trl","name":"trl","category":"training","categoryName":"Training & fine-tuning","audience":["ai-era"],"primary":"ai-era","openClosed":"open","pricing":"open-source","license":"Apache-2.0","maintained":"yes","officialUrl":"http://hf.co/docs/trl","repo":"https://github.com/huggingface/trl","latestVersion":"v1.15.0","latestRelease":"2026-10-08","stars":19565,"starsAsOf":"2026-10-22","description":"Train transformer language models with reinforcement learning.","descriptionZh":"基于 Transformers 的强化学习/偏好训练","callMethods":["SDK","CLI"],"platforms":[],"addedAt":"2026-10-10","updatedAt":"2026-10-22","topics":[],"language":"Python","lastPush":"2026-10-21","formerly":null,"packages":[{"registry":"pypi","name":"trl","url":"https://pypi.org/project/trl/","versionsUrl":"https://pypi.org/project/trl/#history","latestVersion":"1.15.0","latestDate":"2026-10-08"}],"recentVersions":[{"version":"v1.15.0","date":"2026-10-08","prerelease":false,"url":"https://github.com/huggingface/trl/releases/tag/v1.15.0","source":"GitHub Release"},{"version":"v1.14.2","date":"2026-10-06","prerelease":false,"url":"https://github.com/huggingface/trl/releases/tag/v1.14.2","source":"GitHub Release"},{"version":"v1.14.1","date":"2026-09-29","prerelease":false,"url":"https://github.com/huggingface/trl/releases/tag/v1.14.1","source":"GitHub Release"},{"version":"v1.14.0","date":"2026-09-25","prerelease":false,"url":"https://github.com/huggingface/trl/releases/tag/v1.14.0","source":"GitHub Release"},{"version":"v1.13.0","date":"2026-09-10","prerelease":false,"url":"https://github.com/huggingface/trl/releases/tag/v1.13.0","source":"GitHub Release"}],"alternatives":["pytorch","tensorflow","transformers","unsloth","llamafactory","keras","nanogpt","nanochat"],"stacks":["fine-tune-open-models"],"scenarios":[],"problems":[],"reasonZh":"开源大库收录：基于 Transformers 的强化学习/偏好训练","sources":["oss","github-pool"],"record":"https://aimap.page/api/v1/software/trl.json","pages":{"en":"https://aimap.page/en/software/trl/","zh-hans":"https://aimap.page/zh-hans/software/trl/"}}