MiMo v2.6 Pro and v2.6 Flash now on xKiro — omni-modal with a one-million-token context
Omni-modal reasoning over a million-token context: images, audio and video read in the same pass as text. Four MiMo models, native tool calling on every one.
Xiaomi's MiMo line is now served on xKiro across four models. MiMo v2.6 Pro and v2.6 Flash are omni-modal: they read images, audio and video alongside text and answer with extended reasoning in a single pass, over a context window of just over one million tokens and up to 65,536 output tokens. MiMo v2.5 stays available with the same omni-modal reach, while MiMo v2.5 Pro remains the text-only reasoning model for long-horizon software work. All four support native tool calling.
Models in this offer
xiaomi/mimo-v2.6-pro
xiaomi/mimo-v2.6-flash
xiaomi/mimo-v2.5-pro
xiaomi/mimo-v2.5
How to use it
Create an account, open the dashboard and follow the steps shown there. Everything works through the same OpenAI- and Anthropic-compatible endpoints you already use.
Command A+ with vision, tools and a 436K context, North Mini Code for agentic coding, dedicated translation models and the multilingual Aya family. Free — no card needed.
xiaomi/mimo-v2.6-pro
