The new MOPD-based solution reduced repetition rates significantly at a cost of approximately $90,000, compared to an estimated $2.31 million for a full MixRL restart.
Xiaomi has released a real-time post-training dashboard for Mimo 2.6, with the Pro model recording a score of 65.78 on DeepSWE v1.1.
Xiaomi's MiMo team has unveiled MiMo Code, an open-source coding agent designed for complex, long-horizon coding tasks.
Xiaomi collaborated with TileRT to achieve a generation speed of over 1000 tokens per second on a model with 1 trillion parameters.
The MiMo V2.5 series maximizes long-context processing and multimodal performance through Hybrid SWA and MoE.
Xiaomi's MiMo model demonstrated balanced creative writing performance, combining both logic and emotional sensitivity in equal measure.
Xiaomi has released the Mimo V2 Flash evaluation framework to measure the balance between helpfulness and safety in LLMs.