阿里 Qwen 团队发布 QwenGyre：面向超长程智能体的弹性强化学习框架
Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon Agents
Qwen 团队放出了 QwenGyre，一个训练编程智能体的弹性 RL 框架，能跑几小时的 rollout、处理近百万 token，值得做智能体训练的开发者看看。
Alibaba 的 Qwen 团队推出 QwenGyre，一个用于极端长程智能体训练的弹性强化学习（RL）框架。该系统能通过未修改的黑箱 harness 训练编程智能体，支持单次 rollout 运行数小时、处理接近 100 万 token 的任务。QwenGyre 已被应用于 Qwen 的编程智能体训练流程。
Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon Agents
Alibaba’s Qwen team has introduced QwenGyre, an elastic reinforcement learning RL framework for extreme-long-horizon agent training. QwenGyre is an online reinforcement learning system from Alibaba that trains coding agents through unmodified, black-box harnesses. It targets rollouts that run for hours and process close to 1M tokens. The QwenGyre elastic RL framework was used on Qwen’s […] The post Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon Agents appeared first on MarkTechPost .
