# TRL v1.13 发布：长上下文训练

- 来源：Thomas Wolf (@Thom_Wolf)
- 发布时间：2026-09-10 16:43
- AIHOT 分数：45
- AIHOT 链接：https://aihot.news/items/cmtvb3d5j084erok9ja69km2r
- 原文链接：https://x.com/Thom_Wolf/status/2097969147967586632

## AI 摘要

TRL v1.13 发布！我们的开源 RL 训练库，用于对基础模型进行后训练

本次新版本聚焦"长上下文训练"，新增了一份关于如何用 1M+ token 上下文对模型进行后训练的指南 https://huggingface.co/docs/trl/long_context_training ，以及一如既往的速度和内存使用方面的多项改进

详见 https://github.com/huggingface/trl

## 正文

TRL v1.13 is out! our open-source RL training library to to post-train foundation models

this new release is focusing on "long context training" with a new guide on how to post-train model with 1M+ token context https://huggingface.co/docs/trl/long_context_training + various improvements on speed and memory usage as usual

check it out at https://github.com/huggingface/trl

### 引用推文

> Quentin Gallouédec：1M token is basically the entire Harry Potter series ⚡️🪄