The LLM benchmark aggregator.

GPT-4o Mini

OpenAI

Strongest in Multimodal understanding (#6 of 19), weakest in Agents (#227 of 268). Above par in 7 of 26 scopes. Among the models it meets almost everywhere, it finishes behind Claude Sonnet 4 and Claude Opus 4.7 and ahead of Mistral Large and Llama 4 Scout Instruct.

§ 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.

Safety53−8.5#103/3372/3

Safe-prompt compliance54.2−6.2#40/821/1

Jailbreak resistance58.1−6.8#67/2721/1

Toxicity avoidance56.1−2.9#74/2721/1

Secure code57.3−9.2#92/2741/1

Harm refusal53.1−8.5#108/3002/2

Fairness46.4−23.6#161/3002/2

Knowledge42.1−21.4#115/1382/2

Multimodal understanding57.3−14.8#6/191/1

Academic knowledge26.9−37#119/1231/1

Professional42.9−18.7#144/1681/1

Medical44.1−22.7#115/1401/1

Finance39.3−23.6#132/1511/1

Coding37.8−31.3#161/1652/5

Code generation35.1−34.9#70/771/2

Agentic coding39.4−29.3#146/1571/4

Core abilities41.9−25.5#177/2041/3

General intelligence38.3−31.8#184/2041/3

Reasoning32.6−34.3#178/1783/4

Science27.8−35.7#114/1221/1

Mathematics35.4−30.3#129/1401/2

Reasoning33−36.1#137/1402/3

Human preference49.1−18.4#206/3421/1

Agents41.4−26.7#227/2681/5

Knowledge work37.1−36.5#147/1781/2

§ 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 GPT-4o Mini, left for the other.

Command R Pluswon 14 · lost 2

Gemini 1.5 Flashwon 10 · lost 6

Mistral Smallwon 11 · lost 8

Llama 4 Scout Instructwon 12 · lost 10

Mistral Largewon 9 · lost 8

Grok Build 0.1won 8 · lost 8

Grok 4.3won 12 · lost 12

Nemotron 3 Ultra 550B A55Bwon 7 · lost 9

Command Awon 9 · lost 12

GPT-4.1 Nanowon 10 · lost 14

Ling 3.0 Flashwon 7 · lost 10

Grok 4.1 Fastwon 8 · lost 12

Kimi K2.7 Codewon 6 · lost 10

Gemini 1.5 Prowon 7 · lost 12

Mimo V2.6 Prowon 6 · lost 12

DeepSeek V3won 8 · lost 17

DeepSeek R1won 7 · lost 17

MiniMax M3won 7 · lost 17

GLM-4.5won 5 · lost 13

Gemini 2.0 Flashwon 6 · lost 16

GLM-4.6won 6 · lost 16

Kimi K2 Thinkingwon 5 · lost 14

Grok 3won 5 · lost 14

Gemma 4 31Bwon 5 · lost 15

Llama 4 Maverick Instructwon 6 · lost 18

DeepSeek V4 Flashwon 6 · lost 18

DeepSeek V4 Prowon 6 · lost 18

Gemini 3.6 Flashwon 6 · lost 18

GPT OSS 20Bwon 5 · lost 16

Grok 4won 5 · lost 18

Kimi K2.5won 5 · lost 18

Claude 3.5 Haikuwon 4 · lost 15

Kimi K2.6won 5 · lost 19

GLM-5.3 Flashwon 5 · lost 19

GLM-5.2won 5 · lost 19

Gemini 2.5 Prowon 5 · lost 20

GPT-6 Solwon 4 · lost 17

Grok 4 Fastwon 3 · lost 13

Mistral Medium 3.5won 3 · lost 13

O3 Miniwon 4 · lost 19

DeepSeek V3.2won 4 · lost 19

Gemini 2.5 Flashwon 4 · lost 20

Grok 4.5won 4 · lost 20

Qwen3.7 Maxwon 4 · lost 20

Gemini 3.5 Flashwon 4 · lost 20

Gemini 3.8 Flashwon 4 · lost 20

GPT-6 Lunawon 3 · lost 18

GLM-5.1won 3 · lost 18

Claude 3.5 Sonnetwon 3 · lost 19

Mimo V2.5 Prowon 3 · lost 19

Kimi K2won 3 · lost 20

GPT OSS 120Bwon 3 · lost 21

GPT-5.4won 3 · lost 21

Qwen3.8 Maxwon 3 · lost 21

GPT-5.6 Solwon 3 · lost 22

Claude Opus 5won 3 · lost 22

Gemini 2.5 Flash Litewon 2 · lost 15

Gemini 3.1 Flash Litewon 2 · lost 15

Mistral Large 3won 2 · lost 16

GPT-5.4 Miniwon 2 · lost 16

Gemini 3.5 Flash Litewon 2 · lost 16

Gemini 3 Prowon 2 · lost 18

Claude Opus 4.6won 2 · lost 18

GPT-5.1won 2 · lost 20

Claude Sonnet 4.6won 2 · lost 20

GPT-4.1won 2 · lost 21

Qwen3 235B A22B Instructwon 2 · lost 21

GPT-4.1 Miniwon 2 · lost 22

Claude Haiku 4.5won 2 · lost 22

Grok 4.6won 2 · lost 22

GPT-5.6 Lunawon 2 · lost 22

GPT-5.6 Terrawon 2 · lost 22

GPT-5.5won 2 · lost 22

Muse Spark 1.1won 2 · lost 22

GLM-5won 1 · lost 15

GPT-6 Astrawon 1 · lost 15

Qwen3.6 Pluswon 1 · lost 17

Claude Opus 4.5won 1 · lost 17

DeepSeek V4.1 Flashwon 1 · lost 17

GPT-5.4 Nanowon 1 · lost 18

Gemini 3.1 Prowon 1 · lost 18

Kimi K3won 1 · lost 18

O1won 1 · lost 19

GPT-5 Nanowon 1 · lost 20

Claude 3.7 Sonnetwon 1 · lost 20

Claude Opus 5.5won 1 · lost 20

Claude Opus 4won 1 · lost 22

GPT-5.2won 1 · lost 22

GPT-5 Miniwon 1 · lost 23

O4 Miniwon 1 · lost 23

O3won 1 · lost 23

GPT-5won 1 · lost 23

Claude Sonnet 4.5won 1 · lost 23

Claude Opus 4.8won 1 · lost 23

Muse Spark 1.2won 0 · lost 17

Qwen3.8 27Bwon 0 · lost 18

GLM-5.3won 0 · lost 18

Gemini 3.7 Flashwon 0 · lost 18

Claude Fable 5.1won 0 · lost 18

Claude Fable 5won 0 · lost 18

MiniMax M2.7won 0 · lost 20

Grok 4.7won 0 · lost 21

GPT-4owon 0 · lost 22

Inklingwon 0 · lost 24

Claude Sonnet 5won 0 · lost 24

Claude Opus 4.7won 0 · lost 24

Claude Sonnet 4won 0 · lost 25

§ 3 · Sources

Where the numbers come from

11 publications, 26 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26

LMArena Text 1318

GDPval-AA ↗1 measureread 2026-09-26

GDPval-AA -37

ARC-AGI-2 ↗1 measureread 2026-09-26

ARC-AGI-2 0%

LiveCodeBench ↗1 measureread 2026-09-26

LiveCodeBench 27.5%

Aider polyglot ↗1 measureread 2026-09-26

Aider polyglot 3.6%

MMMU-Pro ↗1 measureread 2026-09-26

MMMU-Pro 37.6%

Kagi LLM Benchmark ↗1 measureread 2026-09-26

Kagi LLM Benchmark 28.8%

SimpleBench ↗1 measureread 2026-09-26

SimpleBench 10.7%

Vals.ai ↗7 measuresread 2026-09-26

Vals · CorpFin 45.45%Vals · TaxEval 60.55%Vals · MortgageTax 54.49%Vals · MedQA 72.44%Vals · GPQA Diamond 44.19%Vals · MMLU Pro 62.73%Vals · AIME 11.46%

HELM Safety ↗5 measuresread 2026-09-26

HELM Safety · HarmBench 84.9%HELM Safety · SimpleSafetyTests 97.8%HELM Safety · Anthropic Red Team 98.3%HELM Safety · BBQ 88.2%HELM Safety · XSTest 96%

Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26

Enkrypt · Jailbreak risk 5.9%Enkrypt · Harmful content risk 39.4%Enkrypt · CBRN risk 8%Enkrypt · Toxicity risk 2%Enkrypt · Bias risk 86.3%Enkrypt · Insecure code risk 18.7%

Badge

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