LLM ReferenceLLM Reference

Xiaomi MiMo-V2.5

ProprietaryMultimodal

About

Xiaomi MiMo-V2.5 was released on April 22, 2026 as a native omnimodal agent foundation model with visual and audio understanding, tool-use-oriented post-training, and up to 1M tokens of context. Xiaomi positions it as a new model that surpasses MiMo-V2-Pro in agentic performance and improves multimodal perception over MiMo-V2-Omni; OpenRouter lists xiaomi/mimo-v2.5 at $0.40/M input and $2/M output tokens.

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution

Providers(1)

ProviderInput (per 1M)Output (per 1M)Type
OpenRouter$0.4$2Serverless

Rankings

Best Multimodal / Vision LLMsBest LLMs for Reasoning & MathBest LLMs for Function Calling & Tool UseCheapest LLM APIsBest Long Context LLMsBest LLM APIsBest LLMs for EnterpriseBest LLMs for WritingBest LLMs for Customer Support

Specifications

FamilyMiMo
Released2026-04-22
Context1M
Specializationomnimodal agent
LicenseProprietary
Trainingpretraining

Created by

Xiaomi

Consumer electronics and AI research.

Beijing, China
Founded 2010
Website

Providers(1)

OpenRouter

Links

Website