Request a Pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Ministral 3 3B

Mistral · 3.8B · open weights

Strongest in Factual grounding (#104 of 104), weakest in Safety (#305 of 327). Above par in 0 of 4 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and GPT-6 Astra and ahead of Magistral Small 1.2 and GPT-4.

§ 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.

Agents43.4#242/298
Knowledge work36.9#170/208
Safety42.5#305/327
Factual grounding26#104/104

§ 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 Ministral 3 3B, left for the other.

Llama 3.1 8B Instructwon 2 · lost 1
O3 Miniwon 2 · lost 1
Gemma 4 E2Bwon 2 · lost 1
GPT-4.1 nanowon 2 · lost 1
Gemma 4 E4Bwon 2 · lost 1
GPT-4o miniwon 2 · lost 1
GPT-4won 2 · lost 1
Magistral Small 1.2won 2 · lost 1
DeepSeek V3won 2 · lost 2
Ministral 3 8Bwon 2 · lost 2
Phi-4 Mini Instructwon 2 · lost 2
NVIDIA Nemotron 3 Nano 30B A3Bwon 2 · lost 2
Ministral 3 14Bwon 2 · lost 2
Gemma 3 27B ITwon 2 · lost 2
DeepSeek R1won 2 · lost 2
Gemma 3 12B ITwon 2 · lost 2
Llama 4 Scout Instructwon 2 · lost 2
Llama 3.3 70B Instructwon 2 · lost 2
GPT-4owon 2 · lost 2
Llama 4 Maverick Instructwon 2 · lost 2
Inklingwon 1 · lost 2
Ministral 8Bwon 1 · lost 2
Grok 4.1 Fastwon 1 · lost 2
Gemma 4 31Bwon 1 · lost 3
Trinity Largewon 1 · lost 3
Gemini 3.1 Flash-Litewon 1 · lost 3
Mistral Large 3won 1 · lost 3
Mercury 2won 1 · lost 3
Gemini 2.5 Prowon 1 · lost 3
Qwen3 14Bwon 1 · lost 3
Qwen3 8Bwon 1 · lost 3
Gemma 3 4B ITwon 0 · lost 3
Claude 3.5 Haikuwon 0 · lost 3
Amazon Nova Microwon 0 · lost 3
GPT OSS 20Bwon 0 · lost 3
Amazon Nova Prowon 0 · lost 3
Gemma 4 12Bwon 0 · lost 3
Qwen3.5 397B A17Bwon 0 · lost 3
NVIDIA Nemotron 3 Super 120B A12Bwon 0 · lost 3
Phi 4won 0 · lost 3
GPT-6 Lunawon 0 · lost 3
Gemini 3.5 Flashwon 0 · lost 3
Qwen3 4B Instructwon 0 · lost 3
GPT-4.1 miniwon 0 · lost 3
Grok 4.3won 0 · lost 3
Mistral Large 4won 0 · lost 3
Gemini 2.5 Flash-Litewon 0 · lost 3
Ling 3.0 Flashwon 0 · lost 3
Mistral Large 2.1won 0 · lost 3
Gemini 3.8 Flashwon 0 · lost 3
NVIDIA Nemotron 3 Ultra 550B A55Bwon 0 · lost 3
Kimi K2.7 Codewon 0 · lost 3
MiMo V2.5 Prowon 0 · lost 3
Hy3won 0 · lost 3
Grok 4.5won 0 · lost 3
Qwen3.7 Maxwon 0 · lost 3
Grok Build 0.1won 0 · lost 3
GLM-5.1won 0 · lost 3
Ling 3.0 Flash VLwon 0 · lost 3
Command Awon 0 · lost 3
GLM-5.3 Flashwon 0 · lost 3
Qwen3.8 Maxwon 0 · lost 3
Gemini 3.6 Flashwon 0 · lost 3
GPT-5.6 Terrawon 0 · lost 3
GPT-5 nanowon 0 · lost 3
Qwen3 235B A22B Instructwon 0 · lost 3
Grok 4.6won 0 · lost 3
O4 Miniwon 0 · lost 3
GPT-4.1won 0 · lost 3
GPT-5.2won 0 · lost 3
Gemini 2.5 Flashwon 0 · lost 3
GLM-5.2won 0 · lost 3
Muse Spark 1.1won 0 · lost 3
DeepSeek V4 Flashwon 0 · lost 3
Kimi K2won 0 · lost 3
Claude Opus 4.8won 0 · lost 3
MiniMax M3won 0 · lost 3
GPT-5.6 Lunawon 0 · lost 3
Claude Sonnet 5won 0 · lost 3
MiMo V2.6 Prowon 0 · lost 3
Grok 4.7won 0 · lost 3
MiMo V2.6 Flashwon 0 · lost 3
Gemini 3 Prowon 0 · lost 3
Claude Opus 4.5won 0 · lost 3
Claude Opus 5won 0 · lost 3
Gemma 4 26B A4Bwon 0 · lost 4
Mistral Medium 3.1won 0 · lost 4
Qwen3 Next 80B A3B Thinkingwon 0 · lost 4
Amazon Nova 2 Litewon 0 · lost 4
Gemini 3.1 Prowon 0 · lost 4
Qwen3.5 122B A10Bwon 0 · lost 4
Qwen3 32Bwon 0 · lost 4
GPT OSS 120Bwon 0 · lost 4
Qwen3.5 35B A3Bwon 0 · lost 4
GPT-5.4 nanowon 0 · lost 4
Claude Sonnet 4won 0 · lost 4
MiniMax M2.7won 0 · lost 4
GPT-5.4 miniwon 0 · lost 4
GPT-5 miniwon 0 · lost 4
GPT-5.1won 0 · lost 4
GPT-5won 0 · lost 4
GLM-4.7won 0 · lost 4
Kimi K2.5won 0 · lost 4
Claude Haiku 4.5won 0 · lost 4
DeepSeek V3.2won 0 · lost 4
GPT-5.4won 0 · lost 4
Kimi K2.6won 0 · lost 4
GPT-5.5won 0 · lost 4
Claude Opus 4.7won 0 · lost 4
Claude Sonnet 4.5won 0 · lost 4
DeepSeek V4 Prowon 0 · lost 4
Claude Sonnet 4.6won 0 · lost 4
DeepSeek V4.1 Flashwon 0 · lost 4
GLM-4.6won 0 · lost 4
GPT-5.6 Solwon 0 · lost 4
GPT-6 Solwon 0 · lost 4
Gemini 4 Argonwon 0 · lost 4
GPT-6 Astrawon 0 · lost 4
Claude Opus 5.5won 0 · lost 4

§ 3 · Sources

Where the numbers come from

2 publications, 2 figures. Every one links to the page it was read from.

GDPval-AA ↗1 measureread 2026-11-12
GDPval-AA 37
Vectara Hallucination Leaderboard ↗1 measureread 2026-11-12
Vectara · Factual consistency 75.8%

Badge

PublicAI Index badge for Ministral 3 3B[![PublicAI Index](https://publicai.io/model-index/badge?model=ministral-3-3b)](https://publicai.io/model-index/m/ministral-3-3b)