MiMo-V2.5

xiaomi/mimo-v2.5-20260422

Xiaomi MiMo-V2.5 is a model from Xiaomi released on 2026-04-22 that accepts text, audio, image, and video inputs and produces text output. It supports tools and reasoning, provides open weights, and has a context length of 1,048,576 tokens. Pricing is $0.14 per million input tokens and $0.28 per million output tokens.

  • tool use
  • structured outputs
  • reasoning
  • open-weight

At a glance

providerxiaomi
releasedApr 22, 2026
context1.1M
input / M$0.14

Specifications

What the API gives you

Context window
1.1M tokens
Max output
131K tokens
Input price
$0.14 / 1M tokens
Output price
$0.28 / 1M tokens
Cache read
$0.003 / 1M tokens
Modalities
text, audio, image, video → text
Knowledge cutoff
Released
Apr 22, 2026

Recommendation coverage

Where this model appears

Not currently included in a published recommendation. This means the available task-specific evidence did not support a ranking yet—not that the model was reviewed and rejected.

Measured evidence

Public benchmark record

LMArena Text
1426 Elo2026-07-16
LMArena Vision
1254 Elo2026-07-12
LMArena WebDev
1429 Elo2026-07-16
LMArena WebDev
1427 Elo2026-07-20

LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open.

Open-weight model on Hugging Face.

Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.