XiaomiMiMo/MiMo-V2.6-Pro-RL — new model trending #30 on Hugging Face
Xiaomi released flagship MiMo-V2.6-Pro-RL, a 1.02T omnimodal MoE activating 42B parameters with 1M context.
Xiaomi released MiMo-V2.6-Pro-RL, the flagship checkpoint of the MiMo-V2.6 series, on Hugging Face. It is a sparse mixture-of-experts model with 1.02 trillion total parameters, 42 billion activated, a 1 million-token context, and native text, image, video, and audio support. Xiaomi reports 71.9 on DeepSWE v1.1, 89.9 on Terminal Bench 2.1, and 80.2 on MiMo Cyber Bench. On ExploitBench it reports 47.9, below the 78.5 it lists for GPT-5.6 Sol.
- Flagship sparse MoE has 1.02T total and 42B activated parameters.
- Context is 1M tokens across text, image, video, and audio.
- Xiaomi reports 71.9 DeepSWE and 80.2 on MiMo Cyber Bench.
- ExploitBench score is 47.9, versus 78.5 reported for GPT-5.6 Sol.
- Groupwise grading and MOPD2 distillation follow the mixed RL run.
- Tags
- xiaomimimo-v2.6mixture-of-expertsreinforcement-learningomnimodalopen-weightscybersecurity-benchmarks
- Vendors
- Xiaomi
- Organizations
- Xiaomi
- AI models
- MiMo-V2.6-Pro-RL
From the original article
6 min read at huggingface.co<source srcset="https://github. com/XiaomiMiMo/MiMo/raw/main/figures/Xiaomi_MiMo_darkmode. png?
By XiaomiMiMo · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.