Skip to content

MiniMax mit Released Feb 12, 2026 Updated

MiniMax M2.5

ToolsGuide Scores, rankings, test results and price.

Model summary

  • Text #165 / 416

    277 / 1,000

    ToolsGuide Score· votes only

  • Coding #96 / 147

    281 / 1,000

    ToolsGuide Score· votes 274, tests 292

  • Price per 1M tokens

    In $0.30· Out $1.20

    $0.30

    Input · median text model $0.43

Summary

MiniMax M2.5 is MiniMax's model, released Feb 12, 2026. It's ranked in 2 categories and does best at text: #165 of 416, with a ToolsGuide Score of 277 against a median of 235.

It also ranks #96 of 147 for coding (281).

Its text score uses votes only, because it has published fewer than two of the tests used for that category.

It costs $0.30 per million input tokens and $1.20 per million output tokens, below the median text model ($0.43).

It has published results on 8 tests and has the top score on 2 of them, including BFCL_v3_MultiTurn and MEWC.

Specifications

Creator
MiniMax
Licence
mit
Released
Feb 12, 2026
Context window
1M tokens
Input price
$0.30 / 1M
Output price
$1.20 / 1M
Tests reported
8
Settings ranked
minimax-m2.5, MiniMax-M2.5

Model comparison

Against the ten best text models.

Text score

ToolsGuide Score out of 1,000 · Higher is better

Coding score

ToolsGuide Score out of 1,000 · Higher is better

Input price

USD per 1M input tokens · Lower is better

Leaderboards

Where MiniMax M2.5 sits among the top 20 in each category.

Text

#165 of 416 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model

Full text leaderboard →

Coding

#96 of 147 · ToolsGuide Score out of 1,000, where 1,000 is a flawless model

Full coding leaderboard →

By task

Scored from votes on each kind of prompt alone, on the same 1,000-point scale.

Score against price

Text · the 10 priced models ranked closest · up and to the left is better value

Against its nearest rivals

Text prompts by kind · ToolsGuide Score from votes · hover for values

Text

minimax-m2.5 · hover a bar for its rank

Coding

MiniMax-M2.5 · hover a bar for its rank

Test results

Every test MiniMax M2.5 has a published result on, out of 100%, with its rank among the models that took it.

  • SWE-Bench Verified

    #13 / 120

    Fixing real bugs in real GitHub projects, checked by the projects' own tests.

    80.2%

    Self-reported

  • BrowseComp

    #27 / 68

    Finding hard-to-locate facts by browsing the web.

    76.3%

    Self-reported · In the agents score

  • SWE-Bench Pro

    #41 / 61

    Harder, longer software engineering tasks from real codebases.

    55.4%

    Self-reported · In the coding score

  • Multi-SWE-Bench

    #2 / 7

    A multilingual benchmark for issue resolving that evaluates Large Language Models' ability to resolve software issues across diverse programming ecosystems. Covers 7 programming languages (Java, TypeScript, JavaScript, Go, Rust, C, and C++) with 1,632 high-quality instances carefully annotated by 68 expert annotators. Addresses limitations of existing benchmarks that focus almost exclusively on Python.

    51.3%

    Self-reported

  • GDPval-MM

    #3 / 3

    GDPval-MM is the multimodal variant of the GDPval benchmark, evaluating AI model performance on real-world economically valuable tasks that require processing and generating multimodal content including documents, slides, diagrams, spreadsheets, images, and other professional deliverables across diverse industries.

    59.0%

    Self-reported

  • BFCL_v3_MultiTurn

    #1 / 2

    Berkeley Function Calling Leaderboard (BFCL) V3 MultiTurn benchmark that evaluates large language models' ability to handle multi-turn and multi-step function calling scenarios. The benchmark introduces complex interactions requiring models to manage sequential function calls, handle conversational context across multiple turns, and make dynamic decisions about when and how to use available functions. BFCL V3 uses state-based evaluation by verifying the actual state of API systems after function execution, providing more realistic assessment of function calling capabilities in agentic applications.

    76.8%

    Unverified

  • VIBE-Pro

    #2 / 2

    VIBE-Pro is an advanced version of the VIBE (Visual & Interactive Benchmark for Execution) benchmark that evaluates LLMs on professional-grade full-stack application development tasks. It measures model performance across complex real-world development scenarios including web, mobile, and backend applications with higher difficulty than the standard VIBE benchmark.

    54.2%

    Unverified

  • MEWC

    #1 / 1

    MEWC is a benchmark that evaluates AI model performance on multi-environment web challenges, testing agents' ability to navigate and complete complex tasks across diverse web environments.

    74.4%

    Unverified