Trending storyPast story
Xiaomi open-sources MiMo-V2.6, using RL to let models teach themselves
1 report1 sourceupdated 8 days ago
What happened
Summary
小米发布了 MiMo-V2.6 系列并开源，核心思路是扩展强化学习规模，让模型在训练中自我提升，不再完全依赖人工标注。不过原文被微信屏蔽，正文没披露具体参数量、跑分和开源仓库链接，所以实际效果和可用性暂时没法判断。如果真能靠 RL 让模型自己迭代，训练成本可能降不少，但这点先别太激动，等 repo 出来再看。
Coverage
Follow the reports to see the story from different sides.
Sep 22
- AI HOT (Curated Pool)Xiaomi open-sources MiMo-V2.6, using RL to let models teach themselves
Xiaomi released and open-sourced the MiMo-V2.6 series, focusing on scaling reinforcement learning for self-improvement. The post is blocked by WeChat and does not disclose specific parameters, performance, or repo links.