The LLM benchmark aggregator.

MiniMax M3.1 Flash

MiniMax

Strongest in Computer use (#11 of 81, on 1 of its 2 boards), weakest in Professional (#108 of 188). Above par in 9 of 15 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Muse Spark 1.2 and Qwen3.8 Flash.

§ 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.

Agents57.2−8.7#30/2893/6

Computer use61.5−6.2#11/811/2

Web research51.6−11#26/491/2

Long-horizon work47.1−26.9#28/481/1

Knowledge work63.7−8.1#29/1982/2

Coding54.3−12.1#42/1762/6

Agentic coding54.6−12.4#40/1692/5

Core abilities54.6−10.6#58/2252/4

Data analysis48.7−17.4#34/481/1

General intelligence58.5−10.3#37/2142/3

Reasoning49.1−15.8#104/1901/5

Reasoning48.5−18.2#75/1441/3

Professional49−14.1#108/1882/2

Document design50.1−17.2#29/481/1

Legal47.6−18.3#103/1641/1

§ 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 M3.1 Flash, left for the other.

GPT-4o miniwon 9 · lost 0

Mistral Medium 3.5won 9 · lost 0

Mercury 2.5won 9 · lost 0

GPT-5.4 miniwon 10 · lost 0

GPT-4.1 nanowon 10 · lost 0

GPT OSS 120Bwon 10 · lost 0

Nemotron 3.5 Super VLwon 14 · lost 0

NVIDIA-Nemotron-3.5-Lightning-30B-A3B (NVFP4)won 15 · lost 0

Claude Haiku 4.5won 15 · lost 0

Gemma 4 31Bwon 15 · lost 0

NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4)won 15 · lost 0

Beam 1.0won 14 · lost 1

Muse Spark Openwon 13 · lost 1

GPT-5.4 nanowon 9 · lost 1

Llama 4 Maverick Instructwon 9 · lost 1

Gemini 3.5 Flash-Litewon 9 · lost 1

DeepSeek V3.2won 9 · lost 1

Qwen3.7 Pluswon 8 · lost 1

GPT-4.1 miniwon 8 · lost 2

Mistral Large 3won 8 · lost 2

Inkling Smallwon 8 · lost 2

DeepSeek V3won 8 · lost 2

MiniMax M3won 12 · lost 3

Inklingwon 12 · lost 3

Kimi K2won 7 · lost 2

GPT-4.1won 7 · lost 2

Gemini 2.5 Flashwon 7 · lost 2

Gemini 3.1 Flash-Litewon 7 · lost 2

Mistral Large 4won 11 · lost 4

Claude Sonnet 4.5 (High, 32K, 20250929)won 8 · lost 3

Claude Sonnet 4won 8 · lost 3

DeepSeek R1won 7 · lost 3

Claude Sonnet 4.6won 10 · lost 5

GLM-5.3 Flashwon 10 · lost 5

Qwen3.8 27Bwon 10 · lost 5

DeepSeek V4 Flash Visionwon 9 · lost 5

Kimi K2.5won 7 · lost 4

GLM-5.2won 6 · lost 4

GPT-5.6 Lunawon 9 · lost 6

GPT-6 Lunawon 9 · lost 6

Gemini 3.8 Flashwon 9 · lost 6

Gemini 3.7 Flashwon 9 · lost 6

Gemini 3.1 Prowon 9 · lost 6

Kimi K2.6won 5 · lost 4

GLM-4.6won 5 · lost 4

Qwen3.8 Flashwon 7 · lost 6

Muse Spark 1.2won 8 · lost 7

DeepSeek V4 Flashwon 5 · lost 5

GPT-5.4won 5 · lost 5

GPT-5 miniwon 5 · lost 5

Gemini 3.5 Flashwon 5 · lost 5

Step 5won 7 · lost 8

GPT-5.2won 4 · lost 5

Qwen3.7 Maxwon 4 · lost 5

O4 Miniwon 4 · lost 5

Gemini 3.6 Flashwon 4 · lost 5

DeepSeek-V4-Pro (High, 20260813)won 4 · lost 6

O3won 4 · lost 6

Gemini 2.5 Prowon 4 · lost 6

GPT-5won 4 · lost 6

Claude Sonnet 5won 6 · lost 9

GPT-5.6 Terrawon 6 · lost 9

Qwen3.8 Maxwon 6 · lost 9

Muse Spark 1.1won 3 · lost 6

Claude Opus 4.5 (High, 32K, 20251101)won 3 · lost 6

GLM-5.3won 5 · lost 10

Kimi K3won 5 · lost 10

GPT-5.5won 5 · lost 10

Claude Opus 4.7won 3 · lost 7

Grok 4.5won 4 · lost 11

GPT-5.6 Solwon 4 · lost 11

Grok 4won 2 · lost 7

Gemini 3 Prowon 2 · lost 7

Claude Haiku 5.5won 3 · lost 12

Muse Spark 1.3won 3 · lost 12

Grok 4.6won 2 · lost 13

Claude Sonnet 5.5won 2 · lost 13

MiMo V2.6 Prowon 2 · lost 13

Claude Opus 4.8won 2 · lost 13

MiMo V2.6 Flashwon 1 · lost 8

GPT-6 Solwon 1 · lost 14

DeepSeek V4.1 Flashwon 1 · lost 14

Grok 4.7won 1 · lost 14

Claude Opus 5.5won 1 · lost 14

Muse Wonderwon 1 · lost 14

Gemini 4 Argonwon 1 · lost 14

GPT-6.1 Solwon 1 · lost 14

GPT-6 Astrawon 1 · lost 14

Muse Wonder Fullwon 0 · lost 9

Claude Opus 5won 0 · lost 15

Claude Fable 5.1won 0 · lost 15

Claude Fable 5won 0 · lost 15

§ 3 · Sources

Where the numbers come from

5 publications, 14 figures. Every one links to the page it was read from.

Artificial Analysis ↗1 measureread 2026-10-21

Artificial Analysis Intelligence Index 43

GDPval-AA ↗1 measureread 2026-10-21

GDPval-AA 1458

AA-Briefcase ↗1 measureread 2026-10-21

AA-Briefcase 1474

LiveBench ↗8 measuresread 2026-10-21

LiveBench 73.1LiveBench · Agentic 34.7LiveBench · Reasoning 82.9LiveBench · Coding 83.9LiveBench · Browser Use 92LiveBench · Data Analysis 71.4LiveBench · Research 72.7LiveBench · Design 74

Vals.ai ↗3 measuresread 2026-10-21

Vals · Legal Research Bench 34.13%Vals · Harvey Legal Agent Benchmark 7.5%Vals · Vibe Code Bench 73.94% (OpenHands)

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=minimax-m3-1-flash)](https://publicai.io/model-index/m/minimax-m3-1-flash)