The LLM benchmark aggregator.

Mistral Small 3.1 24B Instruct

Mistral · 24B · open weights

Strongest in Academic knowledge (#114 of 123), weakest in Agents (#244 of 289). Above par in 0 of 13 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of GPT-4.1 Nano and Laguna M.1.

§ 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.

Knowledge42.1−20.7#166/1901/3

Academic knowledge31−32.9#114/1231/1

Professional42.3−20.8#166/1871/2

Legal43.1−23.3#127/1621/1

Medical42.3−24.2#127/1491/1

Finance40.1−22.4#136/1611/1

Reasoning34.4−30.4#184/1891/5

Science27.8−35.7#115/1221/1

Mathematics33.3−30.7#143/1491/2

Human preference47.5−21.1#225/3461/1

Agents42.9−23#244/2892/6

Knowledge work38.6−33.2#153/1982/2

§ 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 Mistral Small 3.1 24B Instruct, left for the other.

Jamba 1.6 Miniwon 8 · lost 1

Jamba 1.6 Largewon 8 · lost 1

Command R Pluswon 8 · lost 1

NVIDIA Nemotron 3.5 Lightning 30B A3Bwon 7 · lost 3

Laguna M.1won 5 · lost 3

GPT-4.1 Nanowon 8 · lost 5

Laguna XS.2won 4 · lost 4

Gemini 1.5 Flashwon 5 · lost 5

Claude 3.5 Haikuwon 6 · lost 6

Mistral Smallwon 5 · lost 6

Llama 4 Scout Instructwon 5 · lost 8

Mistral Medium 3.5won 5 · lost 8

Mercury 2.5won 3 · lost 5

Magistral Small 1.2won 4 · lost 7

GPT-4o Miniwon 4 · lost 8

Mistral Medium 3.1won 3 · lost 7

Gemini 1.5 Prowon 3 · lost 8

DeepSeek R1won 3 · lost 9

Llama 4 Maverick Instructwon 3 · lost 10

Magistral Medium 1.2won 2 · lost 9

Llama 3.3 70B Instructwon 2 · lost 10

O3 Miniwon 2 · lost 11

DeepSeek V3won 2 · lost 11

GPT-5.4 Nanowon 2 · lost 11

Mistral Large 3won 2 · lost 11

Gemini 3.1 Flash Litewon 2 · lost 11

Mimo V2.5won 2 · lost 11

Gemma 4 31Bwon 1 · lost 9

Qwen3.5 Flashwon 1 · lost 10

GPT-5 Nanowon 1 · lost 11

Command Awon 1 · lost 11

GPT-5.4 Miniwon 1 · lost 11

GPT OSS 20Bwon 1 · lost 12

GPT-4owon 1 · lost 12

GPT OSS 120Bwon 1 · lost 12

Claude Haiku 4.5won 1 · lost 12

NVIDIA Nemotron 3 Ultra 550B A55Bwon 1 · lost 12

MiniMax M3won 1 · lost 12

Qwen3.6 Pluswon 1 · lost 12

Gemini 3.5 Flash Litewon 1 · lost 12

Mimo V2.5 Prowon 1 · lost 12

Gemini 3.1 Prowon 1 · lost 12

Muse Wonder Fullwon 0 · lost 8

DeepSeek V4won 0 · lost 9

Kimi K2 Thinkingwon 0 · lost 9

Qwen 3.5 Pluswon 0 · lost 9

Claude Haiku 5.5won 0 · lost 9

Step 5won 0 · lost 9

Muse Spark 1.3won 0 · lost 9

Muse Wonderwon 0 · lost 9

Ling 3.0 Flashwon 0 · lost 10

Grok 2won 0 · lost 10

Claude 3.5 Sonnetwon 0 · lost 10

Mistral Large 4won 0 · lost 10

Grok 4.7won 0 · lost 10

Mimo V2.6 Flashwon 0 · lost 10

Mimo V2.6 Prowon 0 · lost 10

Gemini 4 Argonwon 0 · lost 10

Mistral Largewon 0 · lost 11

Gemini 2.0 Flashwon 0 · lost 11

Grok 3 Miniwon 0 · lost 11

MiniMax M2.1won 0 · lost 11

MiniMax M2.5won 0 · lost 11

O1won 0 · lost 11

Grok 4 Fastwon 0 · lost 11

GLM-4.5won 0 · lost 11

Grok 3won 0 · lost 11

Qwen3 Maxwon 0 · lost 11

Claude Opus 4won 0 · lost 11

GPT-6 Lunawon 0 · lost 11

Claude Opus 4.1won 0 · lost 11

GLM-5won 0 · lost 11

GPT-6 Solwon 0 · lost 11

Grok 4.20won 0 · lost 11

Gemini 3 Flashwon 0 · lost 11

Claude Sonnet 5.5won 0 · lost 11

GPT-6 Astrawon 0 · lost 11

GPT-6.1 Solwon 0 · lost 11

Claude Opus 4.6won 0 · lost 11

Claude Opus 5.5won 0 · lost 11

Gemini 2.5 Flash Litewon 0 · lost 12

Claude 3.7 Sonnetwon 0 · lost 12

O4 Miniwon 0 · lost 12

Gemini 2.5 Flashwon 0 · lost 12

Grok 4won 0 · lost 12

GPT-4.1won 0 · lost 12

Kimi K2won 0 · lost 12

Qwen3 235B A22B Instructwon 0 · lost 12

Grok 4.1 Fastwon 0 · lost 12

O3won 0 · lost 12

GPT-5.2won 0 · lost 12

Claude Opus 4.5won 0 · lost 12

DeepSeek V4.1 Flashwon 0 · lost 12

Gemini 3 Prowon 0 · lost 12

Muse Spark 1.2won 0 · lost 12

GPT-4.1 Miniwon 0 · lost 13

GPT-5 Miniwon 0 · lost 13

Claude Sonnet 4won 0 · lost 13

Inkling Smallwon 0 · lost 13

MiniMax M2.7won 0 · lost 13

DeepSeek V3.2won 0 · lost 13

GLM-4.6won 0 · lost 13

GPT-5won 0 · lost 13

Qwen3.8 27Bwon 0 · lost 13

DeepSeek V4 Flashwon 0 · lost 13

GLM-4.7won 0 · lost 13

Inklingwon 0 · lost 13

Grok 4.3won 0 · lost 13

Gemini 2.5 Prowon 0 · lost 13

Kimi K2.5won 0 · lost 13

GPT-5.6 Lunawon 0 · lost 13

Grok 4.6won 0 · lost 13

GPT-5.1won 0 · lost 13

Claude Sonnet 4.5won 0 · lost 13

Kimi K2.6won 0 · lost 13

Claude Sonnet 5won 0 · lost 13

DeepSeek V4 Prowon 0 · lost 13

GLM-5.1won 0 · lost 13

Grok 4.5won 0 · lost 13

GPT-5.6 Terrawon 0 · lost 13

Claude Sonnet 4.6won 0 · lost 13

Qwen3.7 Maxwon 0 · lost 13

GLM-5.3 Flashwon 0 · lost 13

GPT-5.4won 0 · lost 13

GLM-5.2won 0 · lost 13

Gemini 3.5 Flashwon 0 · lost 13

GLM-5.3won 0 · lost 13

Claude Opus 4.8won 0 · lost 13

GPT-5.5won 0 · lost 13

Qwen3.8 Maxwon 0 · lost 13

Gemini 3.6 Flashwon 0 · lost 13

GPT-5.6 Solwon 0 · lost 13

Kimi K3won 0 · lost 13

Gemini 3.7 Flashwon 0 · lost 13

Muse Sparkwon 0 · lost 13

Claude Opus 5won 0 · lost 13

Muse Spark 1.1won 0 · lost 13

Gemini 3.8 Flashwon 0 · lost 13

Claude Opus 4.7won 0 · lost 13

Claude Fable 5.1won 0 · lost 13

Claude Fable 5won 0 · lost 13

§ 3 · Sources

Where the numbers come from

4 publications, 11 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-10-19

LMArena Text 1303

GDPval-AA ↗1 measureread 2026-10-19

GDPval-AA 377

AA-Briefcase ↗1 measureread 2026-10-19

AA-Briefcase 313

Vals.ai ↗8 measuresread 2026-10-19

Vals · LegalBench 69.16%Vals · CorpFin 44.17%Vals · TaxEval 58.3%Vals · MortgageTax 61.21%Vals · MedQA 69.1%Vals · GPQA Diamond 44.19%Vals · MMLU Pro 66.02%Vals · AIME 3.54%

Badge

[![PublicAI Index](https://publicai.io/model-index/badge?model=mistral-small-3-1-24b-instruct)](https://publicai.io/model-index/m/mistral-small-3-1-24b-instruct)