The LLM benchmark aggregator.

MiniMax M2.5

MiniMax · 229B · open weights

Strongest in Mathematics (#39 of 140, on 1 of its 2 boards), weakest in Human preference (#133 of 342). Above par in 8 of 16 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Opus 4.7 and ahead of Gemini 3.1 Flash Lite and Mimo V2.5.

§ 1 · Profile

What it is good at

Bars run from 50 — the average of the models each source lists — so right of the line is above par. The middle column is the gap to whoever leads that scope.

Reasoning51.8−15.1#76/1782/4

Mathematics56−9.7#39/1401/2

Science54.2−9.3#62/1221/1

Reasoning43.7−25.4#87/1401/3

Professional49.9−11.7#80/1681/1

Medical54.8−12#41/1401/1

Finance50−12.9#88/1511/1

Legal47−20.1#92/1511/1

Core abilities50.8−16.6#82/2041/3

General intelligence51.2−18.9#84/2041/3

Knowledge48.8−14.7#91/1381/2

Academic knowledge48.6−15.3#84/1231/1

Coding46.7−22.4#102/1651/5

Agentic coding46.2−22.5#96/1571/4

Human preference56.1−11.4#133/3421/1

§ 2 · Head to head

What it beats, and what beats it

The same models turn up scope after scope. Counted once: where both were placed, who finished higher. Bars run right for MiniMax M2.5, left for the other.

Laguna XS.2won 10 · lost 0

Laguna M.1won 10 · lost 0

Mercury 2.5won 10 · lost 0

Command R Pluswon 10 · lost 0

Mistral Largewon 10 · lost 0

Gemini 1.5 Flashwon 10 · lost 0

Grok 2won 11 · lost 0

Claude 3.5 Haikuwon 12 · lost 0

Mistral Smallwon 12 · lost 0

GPT OSS 20Bwon 13 · lost 0

Llama 4 Scout Instructwon 14 · lost 0

GPT-4o Miniwon 15 · lost 0

GPT-4.1 Nanowon 16 · lost 0

Gemini 2.0 Flashwon 15 · lost 1

Command Awon 14 · lost 1

Gemini 2.5 Flash Litewon 12 · lost 1

Gemini 1.5 Prowon 11 · lost 1

GPT OSS 120Bwon 14 · lost 2

Claude Haiku 4.5won 14 · lost 2

Mistral Medium 3.5won 13 · lost 2

GPT-4owon 12 · lost 2

GPT-5 Nanowon 12 · lost 2

Llama 4 Maverick Instructwon 13 · lost 3

GPT-5.4 Nanowon 13 · lost 3

Qwen3.6 27Bwon 8 · lost 2

O3 Miniwon 11 · lost 3

DeepSeek V3won 12 · lost 4

Mistral Large 3won 12 · lost 4

MiniMax M2.1won 10 · lost 4

Claude 3.5 Sonnetwon 9 · lost 4

GPT-4.1 Miniwon 11 · lost 5

Kimi K2won 11 · lost 5

Grok 4.3won 10 · lost 6

Qwen3.6 Pluswon 10 · lost 6

Gemini 3.5 Flash Litewon 10 · lost 6

Nemotron 3 Ultrawon 8 · lost 5

Ling 3.0 Flashwon 6 · lost 4

GPT-5.4 Miniwon 9 · lost 6

DeepSeek V3.2won 9 · lost 7

Mimo V2.5won 7 · lost 6

Gemini 3.1 Flash Litewon 8 · lost 7

Grok 3won 7 · lost 7

GPT-5 Miniwon 8 · lost 8

Grok 3 Miniwon 7 · lost 8

DeepSeek R1won 7 · lost 8

GLM-4.6won 7 · lost 8

Mimo V2.5 Prowon 7 · lost 8

MiniMax M2.7won 6 · lost 7

GPT-4.1won 7 · lost 9

MiniMax M3won 7 · lost 9

Inklingwon 7 · lost 9

Grok 4.20won 6 · lost 8

Claude 3.7 Sonnetwon 6 · lost 8

Gemma 4 31Bwon 5 · lost 7

GPT-6 Lunawon 5 · lost 8

Claude Sonnet 4won 6 · lost 10

Grok 4 Fastwon 6 · lost 10

DeepSeek V4won 4 · lost 7

Kimi K2 Thinkingwon 5 · lost 9

Grok 4.1 Fastwon 5 · lost 9

Qwen3 235B A22B Instructwon 5 · lost 10

Qwen3 Maxwon 5 · lost 10

O4 Miniwon 5 · lost 11

Gemini 2.5 Flashwon 5 · lost 11

Kimi K2.5won 5 · lost 11

Kimi K2.6won 5 · lost 11

Qwen3.7 Maxwon 5 · lost 11

GLM-5.2won 5 · lost 11

GLM-4.5won 4 · lost 9

GLM-4.7won 4 · lost 10

Qwen 3.5 Pluswon 3 · lost 8

Claude Opus 4.1won 3 · lost 9

DeepSeek V4 Flashwon 4 · lost 12

GPT-5.6 Lunawon 4 · lost 12

Claude Sonnet 4.5won 4 · lost 12

Claude Sonnet 4.6won 4 · lost 12

GLM-5.3 Flashwon 4 · lost 12

O1won 3 · lost 11

Inkling Smallwon 3 · lost 11

GLM-5.1won 3 · lost 11

Gemini 3 Flashwon 3 · lost 11

DeepSeek V4.1 Flashwon 3 · lost 12

Claude Opus 4won 3 · lost 13

Qwen3.8 27Bwon 3 · lost 13

Gemini 2.5 Prowon 3 · lost 13

Gemini 3.6 Flashwon 3 · lost 13

Gemini 3.1 Prowon 3 · lost 13

Muse Spark 1.1won 3 · lost 13

Qwen3.5 Flashwon 2 · lost 11

Grok 4.7won 2 · lost 11

Grok 4won 2 · lost 14

Claude Sonnet 5won 2 · lost 14

DeepSeek V4 Prowon 2 · lost 14

Claude Opus 4.5won 2 · lost 14

Gemini 3.5 Flashwon 2 · lost 14

Qwen3.8 Maxwon 2 · lost 14

Mimo V2.6 Prowon 1 · lost 10

GPT-6 Solwon 1 · lost 12

Muse Spark 1.2won 1 · lost 14

O3won 1 · lost 15

Grok 4.6won 1 · lost 15

Grok 4.5won 1 · lost 15

GPT-5.6 Terrawon 1 · lost 15

GPT-5.4won 1 · lost 15

GLM-5.3won 1 · lost 15

GPT-5.6 Solwon 1 · lost 15

Gemini 3.7 Flashwon 1 · lost 15

Kimi K3won 1 · lost 15

Gemini 3.8 Flashwon 1 · lost 15

Claude Opus 4.6won 1 · lost 15

Muse Spark 1.3won 0 · lost 12

GPT-6 Astrawon 0 · lost 13

Muse Sparkwon 0 · lost 13

Claude Opus 5.5won 0 · lost 13

GPT-5.1won 0 · lost 14

GPT-5won 0 · lost 16

GPT-5.2won 0 · lost 16

GLM-5won 0 · lost 16

Claude Opus 4.8won 0 · lost 16

GPT-5.5won 0 · lost 16

Gemini 3 Prowon 0 · lost 16

Claude Opus 5won 0 · lost 16

Claude Fable 5.1won 0 · lost 16

Claude Opus 4.7won 0 · lost 16

Claude Fable 5won 0 · lost 16

§ 3 · Sources

Where the numbers come from

4 publications, 13 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1391

ARC-AGI-2 ↗1 measureread 2026-09-26

ARC-AGI-2 4.9%

Kagi LLM Benchmark ↗1 measureread 2026-09-26

Kagi LLM Benchmark 55.2%

Vals.ai ↗10 measuresread 2026-09-26

Vals · CaseLaw 53.48%Vals · LegalBench 79.96%Vals · CorpFin 59.6%Vals · TaxEval 68.15%Vals · MedQA 92.53%Vals · SWE-bench Verified 74.2% (Mini-SWE-agent)Vals · Vibe Code Bench 14.85% (OpenHands)Vals · GPQA Diamond 82.07%Vals · MMLU Pro 80.09%Vals · AIME 88.75%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=minimax-m2-5)](https://publicai.io/model-index/m/minimax-m2-5)