Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

GPT-4o

OpenAI

Strongest in Multimodal understanding (#1 of 19), weakest in Human preference (#181 of 342). Above par in 14 of 24 scopes. Among the models it meets almost everywhere, it finishes behind Grok 4.7 and Claude Fable 5 and ahead of DeepSeek V3 and Grok 4.1 Fast.

§ 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.

Knowledge55.6#43/138
Multimodal understanding72.1#1/19
Academic knowledge39.2#104/123
Safety55.3#48/337
Safe-prompt compliance57.1#22/82
Toxicity avoidance56.7#52/272
Factual grounding50.9#53/101
Secure code61.3#55/274
Jailbreak resistance59#57/272
Harm refusal54.6#78/300
Fairness52.7#89/300
Professional49.5#86/168
Legal52.5#58/151
Medical52.5#66/140
Finance45.9#110/151
Coding40.5#149/165
Code generation36.2#68/77
Agentic coding43.2#119/157
Reasoning35.2#173/178
Science34.5#105/122
Mathematics35.5#128/140
Reasoning35.3#134/140
Human preference50.8#181/342
Human preference50.8#181/342

§ 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 GPT-4o, left for the other.

Jamba 1.6 Miniwon 15 · lost 0
Jamba 1.6 Largewon 15 · lost 0
GPT-4o Miniwon 22 · lost 0
Command R Pluswon 16 · lost 2
Gemini 1.5 Flashwon 15 · lost 2
Mistral Largewon 14 · lost 3
GPT-4.1 Nanowon 17 · lost 4
Llama 4 Scout Instructwon 16 · lost 4
Mistral Smallwon 13 · lost 4
Ling 3.0 Flashwon 12 · lost 4
Command Awon 14 · lost 6
GPT OSS 20Bwon 12 · lost 6
Gemini 1.5 Prowon 13 · lost 7
Grok 4.3won 13 · lost 8
Claude 3.5 Haikuwon 11 · lost 7
Gemini 2.0 Flashwon 12 · lost 9
Grok 4.1 Fastwon 12 · lost 9
DeepSeek V3won 12 · lost 11
Gemma 4 31Bwon 9 · lost 9
DeepSeek V3.2won 10 · lost 11
MiniMax M3won 10 · lost 11
Kimi K2.5won 10 · lost 11
Mimo V2.5 Prowon 9 · lost 10
Gemini 3.1 Flash Litewon 7 · lost 8
GLM-4.6won 9 · lost 11
GLM-5.1won 9 · lost 11
GPT-4 Turbowon 7 · lost 9
Mistral Large 3won 7 · lost 9
GPT-4.1 Miniwon 9 · lost 12
Grok 4won 9 · lost 12
DeepSeek V4 Flashwon 9 · lost 12
Grok 3won 8 · lost 11
GLM-4.5won 7 · lost 10
GPT OSS 120Bwon 9 · lost 13
DeepSeek R1won 9 · lost 13
Mimo V2.6 Flashwon 6 · lost 9
Mimo V2.6 Prowon 6 · lost 9
O4 Miniwon 9 · lost 14
Grok 4 Fastwon 6 · lost 10
MiniMax M2.7won 7 · lost 12
O3 Miniwon 8 · lost 14
Claude Haiku 4.5won 8 · lost 14
Kimi K2won 8 · lost 14
Kimi K2.6won 8 · lost 14
Qwen3.5 Flashwon 5 · lost 10
Gemini 3.5 Flash Litewon 5 · lost 10
Kimi K2 Thinkingwon 6 · lost 12
GPT-5.6 Lunawon 7 · lost 14
Qwen3.7 Maxwon 7 · lost 14
GLM-5.2won 7 · lost 14
Gemini 3.5 Flashwon 7 · lost 14
Gemini 3.6 Flashwon 7 · lost 14
DeepSeek V4 Prowon 7 · lost 15
GPT-6 Solwon 6 · lost 13
Gemini 2.5 Flash Litewon 5 · lost 11
Claude Sonnet 4won 7 · lost 16
Gemini 2.5 Flashwon 7 · lost 16
GPT-5 Nanowon 6 · lost 14
GLM-5.3 Flashwon 6 · lost 15
Qwen3.8 Maxwon 6 · lost 15
GPT-6 Lunawon 5 · lost 13
Qwen3 235B A22B Instructwon 6 · lost 16
Qwen3.6 Pluswon 4 · lost 11
Gemini 2.5 Prowon 6 · lost 17
GPT-5.4 Miniwon 4 · lost 12
GLM-5won 4 · lost 12
Claude 3.7 Sonnetwon 5 · lost 16
O1won 5 · lost 16
Grok 4.5won 5 · lost 16
GPT-5.4 Nanowon 4 · lost 13
Claude 3.5 Sonnetwon 5 · lost 17
GPT-4.1won 5 · lost 17
GPT-5.6 Solwon 5 · lost 18
Gemini 3 Prowon 4 · lost 15
O1 Miniwon 3 · lost 12
Claude Sonnet 4.6won 4 · lost 16
Claude Opus 4.6won 4 · lost 16
Gemini 3.8 Flashwon 4 · lost 17
Gemini 3 Flashwon 3 · lost 13
GPT-5 Miniwon 4 · lost 18
GPT-5won 4 · lost 18
GPT-5.1won 4 · lost 18
GPT-5.4won 4 · lost 18
GPT-5.5won 4 · lost 18
Claude Opus 5won 4 · lost 18
Gemini 3.1 Prowon 3 · lost 14
Claude Opus 4won 4 · lost 19
Inklingwon 3 · lost 18
Grok 4.6won 3 · lost 18
GPT-5.6 Terrawon 3 · lost 18
GPT-5.2won 3 · lost 19
Claude Sonnet 4.5won 3 · lost 19
Claude Opus 4.7won 3 · lost 19
Qwen3.8 27Bwon 2 · lost 13
DeepSeek V4.1 Flashwon 2 · lost 13
Claude Opus 4.5won 2 · lost 15
Muse Spark 1.1won 2 · lost 19
O3won 2 · lost 20
Kimi K3won 1 · lost 15
Claude Opus 5.5won 1 · lost 17
Claude Sonnet 5won 1 · lost 20
Claude Opus 4.8won 1 · lost 20
GLM-5.3won 0 · lost 15
Gemini 3.7 Flashwon 0 · lost 15
Claude Fable 5.1won 0 · lost 15
Claude Fable 5won 0 · lost 15
Grok 4.7won 0 · lost 18

§ 3 · Sources

Where the numbers come from

10 publications, 27 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1336
ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 0%
LiveCodeBench ↗1 measureread 2026-09-26
LiveCodeBench 29.5%
Aider polyglot ↗1 measureread 2026-09-26
Aider polyglot 18.2%
MMMU-Pro ↗1 measureread 2026-09-26
MMMU-Pro 51.9%
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 17.8%
Vals.ai ↗9 measuresread 2026-09-26
Vals · CaseLaw 59.7%Vals · LegalBench 82.21%Vals · CorpFin 45.92%Vals · TaxEval 74.53%Vals · MortgageTax 57.43%Vals · MedQA 88.16%Vals · GPQA Diamond 53.79%Vals · MMLU Pro 72.56%Vals · AIME 11.88%
HELM Safety ↗5 measuresread 2026-09-26
HELM Safety · HarmBench 82.9%HELM Safety · SimpleSafetyTests 98.5%HELM Safety · Anthropic Red Team 99.1%HELM Safety · BBQ 95.1%HELM Safety · XSTest 97.3%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26
Enkrypt · Jailbreak risk 5.2%Enkrypt · Harmful content risk 32.2%Enkrypt · CBRN risk 6.2%Enkrypt · Toxicity risk 1.6%Enkrypt · Bias risk 79.3%Enkrypt · Insecure code risk 10.7%
Vectara · Factual consistency 90.4%

Badge

PublicAI Index badge for GPT-4o[![PublicAI Index](https://publicai.io/model-index/badge?model=gpt-4o)](https://publicai.io/model-index/m/gpt-4o)