A smaller 9B agent model distilled from a larger Xiaomi model
Made by Xiaomi MiMo
View profile on Hugging Face (opens in a new tab)MiMo-V2.6-Distill-Qwen-9B is a 9-billion-parameter model made by fine-tuning Qwen3.5-9B on data generated by Xiaomi's larger MiMo-V2.6 models, a technique called distillation where a smaller model is trained to imitate a bigger one's outputs rather than being trained on raw internet text from scratch.
It covers coding, general-purpose agent tasks, visual coding, and cybersecurity, and the creator releases it specifically as a starting checkpoint for others to continue training with reinforcement learning, rather than as a finished, most-capable product.
You could use it to…
- Start your own agent-training run from a warmed-up checkpoint
- Fine-tune a 9B model already trained on agent tasks
- Try a smaller, cheaper sibling of Xiaomi’s flagship model
Can I use this?
You'll need A GPU with a recent SGLang build · Setup needed
Worth knowing The creator describes this as a supervised fine-tuning checkpoint meant as a research starting point, not the strongest model in the MiMo-V2.6 family, and its reported evaluation numbers include internal test sets not independently reproducible. MIT licence. BuildTube has not run or verified this model.
Why it's here
Not trending this week
Last seen on the Hugging Face trending list on 28 September 2026. It stays on BuildTube so you can still find it.
- 616 people have liked it on Hugging Face.
- It was downloaded 16,996 times in the last 30 days.
Numbers checked on 6 October 2026; not refreshed since.
Behind it
See the code on Hugging FaceBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?