Xiaomi MiMo-V2 Globally Releases Pro and Omni Models on OpenRouter

OpenRouter

Mar 19, 2026 · Updated Apr 25, 2026

Xiaomi globally released MiMo-V2-Pro and MiMo-V2-Omni, now live on OpenRouter. MiMo-V2-Pro is a 1T-parameter flagship built for agentic scenarios with 1M context, with perceived performance approaching Claude Opus 4.6. MiMo-V2-Omni natively handles image, video, and audio in a unified architecture.

Xiaomi MiMo-V2 launched globally with two frontier models on OpenRouter, a unified API gateway for AI models. MiMo-V2-Pro is Xiaomi's flagship — over 1T parameters, 1M context, optimized for agentic scenarios and production tasks — at $1/M input and $3/M output. MiMo-V2-Omni natively processes image, video, and audio in a unified architecture with visual grounding, tool use, and code execution, at $0.40/M input and $2/M output.

MiMo-V2-Pro ranks among the global top tier on PinchBench and ClawBench, with perceived performance approaching Claude Opus 4.6. It's adaptable to agent frameworks like OpenClaw, designed to serve as the orchestration brain of agent systems. MiMo-V2-Omni's architecture combines strong multimodal perception with agentic capability for complex real-world tasks spanning modalities.

Evaluate MiMo-V2-Pro for agentic pipelines via OpenRouter to test performance at its price point. MiMo-V2-Omni is worth trying for cross-modal tasks — video analysis, audio processing, and visual grounding across modalities.

View the full update on openrouter.ai

OpenRouter

@OpenRouterMar 19

Stealth Model Reveal: Hunter and Healer Alpha are @XiaomiMiMo MiMo-V2-Pro and MiMo-V2-Omni Both models are live now on OpenRouter, and free to use in @OpenClaw via the OpenRouter provider for the next week! https://t.co/6mhyowUgK9

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from OpenRouter →

Keep reading

Xiaomi Launches MiMo-V2.5 Series With 1M Context and Reasoning Tokens

Xiaomi released the MiMo-V2.5 series on OpenRouter, featuring a 1 million token context window and native multimodal support for image and video tasks. The models are specifically architected for long-horizon agentic workflows and coding, offering reasoning-enabled thinking tokens to improve task stability. By delivering pro-level performance at roughly half the typical inference cost, these models lower the economic barrier for deploying autonomous agents at scale.

OpenCode Adds Xiaomi MiMo v2.5 Models to Go for Agentic Coding

OpenCodeApr 24

OpenCode Adds Xiaomi MiMo v2.5 Models to Go for Agentic Coding

OpenCode integrated Xiaomi's MiMo v2.5 and v2.5 Pro models into its Go platform, offering native multimodality and specialized coding intelligence. These agent-centric models provide a 1-million-token context window for complex engineering tasks at the same price point as previous versions.

Xiaomi MiMo Slashes V2.5 API Pricing by 99 Percent

MiMoMay 27

Xiaomi MiMo Slashes V2.5 API Pricing by 99 Percent

Xiaomi permanently reduced MiMo-V2.5 Series API costs by up to 99% and eliminated tiered pricing for long-context inputs. The update uses inference optimizations to provide 5–8x more tokens in subscription plans, making high-volume agentic workflows significantly more affordable.

Arena.ai Ranks Xiaomi MiMo-V2.5 as Top Open Source Coding Model

ArenaApr 30

Arena.ai Ranks Xiaomi MiMo-V2.5 as Top Open Source Coding Model

Arena.ai validated Xiaomi's MiMo-V2.5-Pro as a top-three open-weight model for frontend web development following its official open-source release under the MIT license. The model features a 1-million-token context window and native multimodality, offering a high-performance alternative for commercial agentic workflows.