stealthmodels.

MODEL BENCHMARKS

MiMo-V2.6-Pro: Beach

Xiaomi's flagship MiMo V2.6 multimodal reasoning model, with a million-token context and 128K output. Xiaomi publishes the MIT-licensed Pro-RL checkpoint for local deployment.

DATASHEET

MiMo-V2.6-Pro specs and pricing

A good fit forComplex software engineering, long-running agents and multimodal professional work that needs more than Flash, with released weights for self-hosting at trillion-parameter scale.

MiMo-V2.6-ProOther modelsMedian

PriceDeveloper API, input / output per 1M tokens
$0.435 / $0.87Cheaper than 35 of 43 others
Context windowTokens in one request
1,048,576Joint largest with 18 others
Output speedTokens per second, our measurement
116 tokens/sFaster than 19 of 50 others
Developer
Xiaomi
Released
21 Sep 2026
Family
MiMo V2.6
License
MIT
Open model
Yes
Local model
No
Origin
China
Parameters
1.02T
Active parameters
42B
Max output
131,072 tokens
Input → output
Text, image, video, audio → text
Architecture
Sparse Mixture-of-Experts; OpenRouter maps the route to the released Pro-RL checkpoint. The UltraSpeed edition is a separate route
All prices and providers2

Xiaomi MiMo API; input 0.435 USD/1M; cached read 0.0036/1M; output 0.87/1M

  • OpenRouterObserved 2026-10-23
    Input $0.435Output $0.87Cached read $0.0036

    per 1M tokens

    xiaomi/mimo-v2.6-pro
  • Xiaomi MiMo APIObserved 2026-10-23
    Input $0.435Output $0.87Cached read $0.0036

    per 1M tokens

    Available

    mimo-v2.6-pro
Measurements and sources7

MiMo V2.6 family

  1. MiMo-V2.6-Pro48.1This model
  2. MiMo-V2.6-Flash44.7Compare MiMo-V2.6-Pro and MiMo-V2.6-Flash

StealthMark score

StealthMark measures how well a model’s abilities carry over to original, demanding tasks it could not have trained for: physical and spatial reasoning, planning across many steps, and precise visual work. How StealthMark works explains every benchmark, what it tests and why it is hard.

General Intelligence

The ability to reason, plan, solve problems and understand how things relate and behave. A high score means it thinks hard problems through, from careful analysis to plans that span many steps.

Visual Intelligence

The ability to understand visual intent and turn ideas, descriptions and plans into coherent, effective visuals. A high score means design ideas come out the way you meant them, in illustrations, icons, diagrams or interface visuals.

Overall score

Overall shows how well a model’s abilities carry over to original, demanding tasks, weighting every benchmark into one score from 0 to 100. Estimated means the total includes an estimate for one missing scene.

Judging

Scores reflect how well each result meets the benchmark requirements. Game scores come from the recorded play: the points earned on each level and how each level ended. How StealthMark keeps scores honest.

Games

Each game is a sequence of levels, and earlier moves carry into later levels. A level that was not played is shown as not played and never counts as zero. Play every level of Latch or Foresight yourself to get a score calculated like the models’.

Reasoning efficiency

Good results with little thinking score high. On each task, a model’s result and reasoning tokens are compared with every other model and combined with the Balanced Integration Score (Liesefeld & Janczyk, 2019); 50 is average. Single step shows how much reasoning a model needs to understand and solve one problem, multi-step how efficiently it keeps working on a task by building on its earlier reasoning, as in agentic workflows. Not part of StealthMark.

BENCHMARK RESULTS

Compare results

This link opens the same results in the same order.

StealthMark benchmark results

Reset benchmark preferences?

Return to the default view, clear comparison selections and restore the advertisement.