Skip to content

MiniMax mit Released Mar 18, 2026 Updated

MiniMax M2.7

ToolsGuide Scores, rankings, test results and price.

Model summary

  • Text #136 / 416

    324 / 1,000

    ToolsGuide Score· votes only

  • Coding #88 / 147

    292 / 1,000

    ToolsGuide Score· votes 282, tests 299

  • Price per 1M tokens

    In $0.30· Out $1.20

    $0.30

    Input · median text model $0.43

Summary

MiniMax M2.7 is MiniMax's model, released Mar 18, 2026. It's ranked in 2 categories and does best at text: #136 of 416, with a ToolsGuide Score of 324 against a median of 235.

It also ranks #88 of 147 for coding (292).

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 11 tests and has the top score on 4 of them, including Multi-SWE-Bench and VIBE-Pro.

Specifications

Creator
MiniMax
Licence
mit
Released
Mar 18, 2026
Context window
205K tokens
Input price
$0.30 / 1M
Output price
$1.20 / 1M
Tests reported
11
Settings ranked
minimax-m2.7, MiniMax-M2.7

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 M2.7 sits among the top 20 in each category.

Text

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

Full text leaderboard →

Coding

#88 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-m2.7 · hover a bar for its rank

Coding

MiniMax-M2.7 · hover a bar for its rank

Test results

Every test MiniMax M2.7 has a published result on, out of 100%, with its rank among the models that took it.

  • SWE-Bench Pro

    #38 / 61

    Harder, longer software engineering tasks from real codebases.

    56.2%

    Self-reported · In the coding score

  • Terminal-Bench 2.0

    #26 / 54

    Getting real work done in a command-line terminal.

    57.0%

    Self-reported

  • SWE-bench Multilingual

    #14 / 49

    A multilingual benchmark for issue resolving in software engineering that covers Java, TypeScript, JavaScript, Go, Rust, C, and C++. Contains 1,632 high-quality instances carefully annotated from 2,456 candidates by 68 expert annotators, designed to evaluate Large Language Models across diverse software ecosystems beyond Python.

    76.5%

    Self-reported

  • Toolathlon

    #29 / 42

    Tool Decathlon is a comprehensive benchmark for evaluating AI agents' ability to use multiple tools across diverse task categories. It measures proficiency in tool selection, sequencing, and execution across ten different tool-use scenarios.

    46.3%

    Self-reported · In the agents score

  • Finance Agent v2

    #28 / 28

    Finance Agent v2 is an agentic financial-analysis benchmark from Vals that evaluates models on real-world finance workflows, measuring their ability to retrieve and reason over financial documents, perform multi-step calculations, and produce accurate analyses.

    27.9%

    Self-reported

  • NL2Repo

    #22 / 25

    NL2Repo evaluates long-horizon coding capabilities including repository-level understanding, where models must generate or modify code across entire repositories from natural language specifications.

    39.8%

    Self-reported · In the coding score

  • Artificial Analysis

    #11 / 13

    Artificial Analysis benchmark evaluates AI models across quality, speed, and pricing dimensions, providing a composite assessment of model capabilities for real-world usage.

    50.0%

    Self-reported

  • Multi-SWE-Bench

    #1 / 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.

    52.7%

    Self-reported

  • VIBE-Pro

    #1 / 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.

    55.6%

    Unverified

  • MLE-Bench Lite

    #1 / 1

    MLE-Bench Lite evaluates AI agents on machine learning engineering tasks, testing their ability to build, train, and optimize ML models for Kaggle-style competitions in a lightweight evaluation format.

    66.6%

    Unverified

  • MM-ClawBench

    #1 / 1

    MM-ClawBench evaluates models on MiniMax's Claw-style agent benchmark, measuring practical agentic task completion quality in real-world OpenClaw usage scenarios.

    62.7%

    Unverified