MiMo-V2.6-Flash-MOPD
MiMo-V2.6-Flash-MOPDXiaomi’s omnimodal 309B MoE for coding and agents, with 15B active, 1M-token context, and MOPD training to reduce repeated tool calls.
Model details
View repositoryMiMo-V2.6-Flash-MOPD is Xiaomi’s efficiency-balanced omnimodal AI model, built on a sparse Mixture-of-Experts architecture with 309B total parameters (15B activated) and a hybrid sliding-window/global-attention design supporting contexts up to 1M tokens. It natively processes text, images, video, and audio through dedicated vision and audio encoders while producing text outputs. Unlike the RL checkpoint, which is trained through large-scale mixed reinforcement learning on verifiable tasks, the MOPD version further distills multiple domain-specialized RL and SFT teachers into the model on-policy. This extends its capabilities to harder-to-verify domains such as long-horizon game development, scientific research, and embodied intelligence, while reducing repetitive tool calls in agentic workflows.
