MiMo-V2.6 is Xiaomi’s open omnimodal model family for long-horizon agent work. Pro and Flash handle text, images, audio and video with 1M context, while Xiaomi is also releasing the technical report, RL environments and training code behind the public post-training run.

MiMo Code is an open-source terminal AI coding agent built on OpenCode. Optimized for long-horizon tasks, it uses an independent checkpoint subagent to manage unbounded context windows, executes sandbox workflows, and evolves via scheduled maintenance.

MiMo-V2-Flash is a 309B MoE (15B active) model by Xiaomi. It is a powerful, efficient, and ultra-fast foundation language model that particularly excels in reasoning, coding, and agentic scenarios, while also serving as an excellent general-purpose assistant for everyday tasks.

The Xiaomi MiMo-V2.5 series introduces V2.5-Pro for complex, long-horizon software engineering and V2.5 for highly efficient, native omnimodal understanding. Both models match frontier performance while drastically reducing token consumption.

MiMo-V2-Pro and MiMo-V2-Omni are Xiaomi’s new agent foundation models. Pro is built for long-chain coding, tool use, and OpenClaw-style workflows, while Omni adds vision and audio to push the same agentic stack into the real world.

Xiaomi's MiMo-Audio is a breakthrough in open-source audio intelligence. Pre-trained on over 100M hours of data, it's the first audio model to show emergent few-shot generalization and In-Context Learning.

Open-source (Apache 2.0) LLM series 'born for reasoning.' Pre-trained & RL-tuned models (like the 7B) match o1-mini on math/code. Base/SFT/RL models released.