NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/MiniMax M2
minimax logo

MiniMax M2

by minimax · Released Oct 2025

Open Source
72.4
avg score
Rank #31
Compare
Better than 90% of all models
Context
205K tokens (~102 books)
Input $/1M
$0.30
Output $/1M
$1.20
Type
text
License
Open Source
Benchmarks
8 tested
Data as of
About

MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning,...

Tested on 8 benchmarks · BenchGecko score 72.4. Top scores: Chatbot Arena Elo — Overall (1343.4%), Chatbot Arena Elo — Coding (1296.6%), OpenCompass — IFEval (90.2%).

Capabilities
coding
74.0
#11 globally
math
79.1
#29 globally
knowledge
57.9
#72 globally
language
90.2
#12 globally
Benchmark Scores
Compare All
Tested on 8 benchmarks · Ranked across 5 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
codingCompare coding →
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

74.0·
mathCompare math →
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

79.1·
knowledgeCompare knowledge →
OpenCompass — MMLU-Pro

OpenCompass MMLU-Pro evaluation. Harder knowledge test with more answer choices.

81.6·
OpenCompass — GPQA-Diamond

OpenCompass evaluation of GPQA Diamond. PhD-level science questions from the hardest subset.

78.7·
OpenCompass — HLE

OpenCompass evaluation of Humanitys Last Exam. Expert-level cross-discipline knowledge test.

13.4·
Quick compare:z-ai logovs GLM 4.5Anthropic logovs Claude Opus 4.8nousresearch logovs Hermes 3 70B Instruct
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · minimax MiniMax M2
Oct 2025
72.4
$0.30/M in205Kctx8 benchmarks
Jan 2026
$0.30/M in66Kctx(-139K)
See the full MiniMax M2 family →
Recently Happened
MiniMax M2 pricing increased 18%
Sep 28, 2026
MiniMax M2 pricing increased 18%
Sep 24, 2026
Similar Models
z-ai logo
GLM 4.5
z-ai
72.3$0.60/1M
Anthropic logo
Claude Opus 4.8
Anthropic
71.8$5.00/1M
nousresearch logo
Hermes 3 70B Instruct
nousresearch
73.2$0.70/1M
Links
Info
minimaxPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@minimax
Source Code
GitHubHugging Face
BenchGecko API
minimax-m2
Specifications
  • Typetext
  • Context205K tokens (~102 books)
  • ReleasedOct 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.002
Available On
minimax logominimax$0.30
Categories
codingmathknowledgelanguage
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
z-ai logoGLM 4.5
72.3
Anthropic logoClaude Opus 4.8
71.8
nousresearch logoHermes 3 70B Instruct
73.2
Muse Spark
73.3
MiniMax M2 is an open-source text AI model by minimax, released in October 2025. It has an average benchmark score of 72.4. Context window: 205K tokens.

Related Models

GLM 4.5 · z-aiClaude Opus 4.8 · AnthropicHermes 3 70B Instruct · nousresearchMuse Spark · UnknownQwen2.5 Coder 32B Instruct · Alibaba Qwen

Benchmarks

Chatbot Arena Elo — OverallChatbot Arena Elo — CodingOpenCompass — IFEvalOpenCompass — MMLU-ProOpenCompass — AIME2025

Related Pages

minimax · Providerminimax · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • MiniMax M2 by minimax. BenchGecko score 72.4, rank 31 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.30 input · $1.20 output per 1M tokens (as of 2026-10-05).
  • Sold by 3 providers (as of 2026-10-05): Minimax (fp8) $0.26 in / $1.02 out · Google $0.30 in / $1.20 out · Novita (fp8) $0.30 in / $1.20 out. Every provider
  • Gecko Tests: Who Are You E (Sometimes says it is Anthropic) · Tokenizer Tax A (36% more tokens outside English).

How to cite · data as of 2026-10-05

MiniMax M2 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/minimax-m2

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP