Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

Gemini 3.1 Pro

Google

Strongest in Science (#1 of 122), weakest in Safety (#190 of 337). Above par in 17 of 24 scopes. Among the models it meets almost everywhere, it finishes behind Claude Opus 5.5 and Claude Fable 5 and ahead of Muse Spark 1.1 and DeepSeek V4.1 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.

Knowledge60.2#5/138
Academic knowledge62.2#4/123
Core abilities59.5#10/204
Instruction following69.7#3/57
Language61.5#12/57
Data analysis58.6#16/57
General intelligence53.9#68/204
Human preference65.4#14/342
Human preference65.4#14/342
Reasoning56.9#26/178
Science63.5#1/122
Reasoning60.2#18/140
Mathematics51.7#73/140
Professional54.5#38/168
Medical60.2#7/140
Finance54.4#55/151
Legal51.1#70/151
Coding46.4#107/165
Code generation48.2#51/77
Agentic coding45.8#100/157
Agents46.8#155/268
Knowledge work45.8#102/178
Safety49.7#190/337
Factual grounding48.9#59/101

§ 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 Gemini 3.1 Pro, left for the other.

Mistral Large 3won 19 · lost 1
GPT-4o Miniwon 18 · lost 1
GPT-4.1 Nanowon 18 · lost 1
Llama 4 Scout Instructwon 17 · lost 1
Gemini 2.0 Flashwon 16 · lost 1
Nemotron 3 Ultrawon 15 · lost 1
Gemini 3.5 Flash Litewon 20 · lost 2
DeepSeek V3won 19 · lost 2
Mistral Medium 3.5won 15 · lost 2
GPT OSS 20Bwon 14 · lost 2
Gemini 3 Flashwon 14 · lost 2
Grok 4.3won 20 · lost 3
Qwen3.7 Maxwon 20 · lost 3
Claude 3.5 Haikuwon 13 · lost 2
Mistral Smallwon 13 · lost 2
Qwen3.6 Pluswon 19 · lost 3
Command Awon 15 · lost 3
Grok 4 Fastwon 15 · lost 3
Inklingwon 19 · lost 4
GPT-5.4 Miniwon 19 · lost 4
GPT-4owon 14 · lost 3
Grok 4.1 Fastwon 14 · lost 3
Qwen3.6 27Bwon 13 · lost 3
MiniMax M2.5won 13 · lost 3
Grok 3won 13 · lost 3
Claude 3.5 Sonnetwon 12 · lost 3
Qwen3 Maxwon 12 · lost 3
GPT OSS 120Bwon 16 · lost 4
Kimi K2.5won 16 · lost 4
GPT-5.4 Nanowon 19 · lost 5
GPT-4.1 Miniwon 15 · lost 4
Gemini 3.1 Flash Litewon 15 · lost 4
MiniMax M3won 18 · lost 5
Mimo V2.5 Prowon 14 · lost 4
GPT-5 Nanowon 13 · lost 4
Gemini 2.5 Flash Litewon 12 · lost 4
MiniMax M2.7won 12 · lost 4
Claude Haiku 4.5won 15 · lost 5
DeepSeek R1won 15 · lost 5
Kimi K2.6won 18 · lost 6
DeepSeek V4 Flashwon 17 · lost 6
Kimi K2won 14 · lost 5
GLM-4.6won 14 · lost 5
Kimi K2 Thinkingwon 11 · lost 4
Qwen3.5 Flashwon 11 · lost 4
O1won 11 · lost 4
Mimo V2.5won 11 · lost 4
Qwen3.8 27Bwon 16 · lost 6
O3 Miniwon 13 · lost 5
DeepSeek V3.2won 14 · lost 6
GLM-5.2won 16 · lost 7
Gemini 3.5 Flashwon 16 · lost 7
Gemini 3.6 Flashwon 16 · lost 7
Grok 3 Miniwon 11 · lost 5
Inkling Smallwon 11 · lost 5
GLM-4.7won 11 · lost 5
Claude Opus 4won 13 · lost 6
Grok 4won 12 · lost 6
GLM-5won 12 · lost 6
GPT-5.6 Lunawon 15 · lost 8
Claude Opus 4.5won 15 · lost 8
GPT-5 Miniwon 13 · lost 7
O4 Miniwon 13 · lost 7
Gemini 2.5 Flashwon 13 · lost 7
GPT-6 Lunawon 13 · lost 7
Gemma 4 31Bwon 11 · lost 6
GLM-5.1won 11 · lost 6
Claude 3.7 Sonnetwon 10 · lost 6
GPT-5.2won 14 · lost 9
Grok 4.5won 14 · lost 9
GLM-5.3 Flashwon 14 · lost 9
Claude Sonnet 4.5won 12 · lost 8
GLM-5.3won 13 · lost 9
Claude Sonnet 4.6won 14 · lost 10
GPT-4.1won 11 · lost 8
Qwen3 235B A22B Instructwon 11 · lost 8
O3won 11 · lost 8
Qwen3.8 Maxwon 13 · lost 10
GPT-5won 11 · lost 9
Grok 4.7won 11 · lost 9
Gemini 3 Prowon 10 · lost 9
Claude Sonnet 4won 11 · lost 10
Gemini 2.5 Prowon 11 · lost 10
DeepSeek V4.1 Flashwon 11 · lost 10
Muse Spark 1.1won 12 · lost 11
DeepSeek V4 Prowon 12 · lost 12
Claude Sonnet 5won 11 · lost 12
GPT-5.4won 11 · lost 13
Claude Opus 4.6won 10 · lost 12
GPT-5.1won 8 · lost 10
Grok 4.6won 10 · lost 13
GPT-5.6 Terrawon 10 · lost 13
Claude Opus 4.8won 10 · lost 13
Muse Spark 1.2won 9 · lost 12
Claude Opus 4.7won 10 · lost 14
Gemini 3.8 Flashwon 9 · lost 14
GPT-6 Solwon 8 · lost 13
Kimi K3won 8 · lost 14
GPT-5.6 Solwon 8 · lost 16
Gemini 3.7 Flashwon 7 · lost 15
GPT-5.5won 7 · lost 17
Muse Sparkwon 4 · lost 11
GPT-6 Astrawon 4 · lost 17
Claude Opus 5won 4 · lost 19
Muse Spark 1.3won 2 · lost 16
Claude Fable 5.1won 2 · lost 20
Claude Fable 5won 2 · lost 20
Claude Opus 5.5won 1 · lost 19

§ 3 · Sources

Where the numbers come from

9 publications, 33 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1487
Artificial Analysis ↗1 measureread 2026-09-26
Artificial Analysis Intelligence Index 30
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 776
AA-Briefcase ↗1 measureread 2026-09-26
AA-Briefcase 453
ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 77.1%
LiveBench ↗8 measuresread 2026-09-26
LiveBench 77LiveBench · Reasoning 84LiveBench · Coding 76.5LiveBench · Agentic Coding 44.1LiveBench · Mathematics 91LiveBench · Data Analysis 78.5LiveBench · Language 85.4LiveBench · Instruction Following 79.1
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 79.6%
Vals.ai ↗18 measuresread 2026-09-26
Vals · Legal Research Bench 20.67%Vals · CaseLaw 64.84%Vals · LegalBench 87.4%Vals · Harvey Legal Agent Benchmark 0%Vals · Finance Agent 42.98%Vals · CorpFin 64.49%Vals · TaxEval 72.88%Vals · MortgageTax 69.4%Vals · MedQA 96.37%Vals · MedCode 59.06%Vals · MedScribe 76.11%Vals · SWE-bench Verified 78.8% (Mini-SWE-agent)Vals · Vibe Code Bench 32.03% (OpenHands)Vals · Code Migration 17.31%Vals · GPQA Diamond 95.45%Vals · MMLU Pro 90.99%Vals · AIME 98.13%Vals · ProofBench 26%
Vectara · Factual consistency 89.6%

Badge

PublicAI Index badge for Gemini 3.1 Pro[![PublicAI Index](https://publicai.io/model-index/badge?model=gemini-3-1-pro)](https://publicai.io/model-index/m/gemini-3-1-pro)