Request a pilot
‹ PublicAI Index

The LLM benchmark aggregator.

GPT-5

OpenAI

Strongest in Secure code (#10 of 274), weakest in Toxicity avoidance (#112 of 272). Above par in 24 of 26 scopes. Among the models it meets almost everywhere, it finishes behind Claude Fable 5 and Claude Fable 5.1 and ahead of Gemini 3.8 Flash and GPT-5.2.

§ 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.

Safety58.8#12/337
Secure code65.4#10/274
Harm refusal59.3#15/300
Jailbreak resistance62.6#17/272
Fairness66.2#20/300
Safe-prompt compliance56.6#23/82
Factual grounding37.1#91/101
Toxicity avoidance54.8#112/272
Professional56.7#20/168
Legal59.2#16/151
Medical58.5#17/140
Finance54.2#56/151
Core abilities56.8#28/204
General intelligence59.8#29/204
Knowledge55.5#44/138
Academic knowledge56.6#40/123
Reasoning53.7#53/178
Mathematics57.2#23/140
Science56.6#46/122
Reasoning49.2#64/140
Coding52.3#57/165
Agentic coding52.6#55/157
Human preference60.4#83/342
Human preference60.4#83/342
Agents52.1#94/268
Knowledge work53.2#62/178

§ 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-5, left for the other.

MiniMax M2.5won 16 · lost 0
Gemini 1.5 Flashwon 17 · lost 0
Mistral Medium 3.5won 17 · lost 0
Grok 4 Fastwon 18 · lost 0
Gemini 1.5 Prowon 19 · lost 0
GLM-4.5won 19 · lost 0
Kimi K2 Thinkingwon 20 · lost 0
GPT OSS 20Bwon 22 · lost 0
Gemini 2.0 Flashwon 23 · lost 0
GPT-4.1 Nanowon 25 · lost 0
GPT-4.1 Miniwon 24 · lost 1
GPT-4o Miniwon 23 · lost 1
Grok 3won 20 · lost 1
Mistral Smallwon 19 · lost 1
GPT-5.4 Nanowon 19 · lost 1
Mistral Large 3won 19 · lost 1
Gemini 3.1 Flash Litewon 18 · lost 1
Ling 3.0 Flashwon 17 · lost 1
Command R Pluswon 17 · lost 1
GPT OSS 120Bwon 24 · lost 2
DeepSeek V3won 24 · lost 2
DeepSeek V3.2won 23 · lost 2
Llama 4 Scout Instructwon 22 · lost 2
GLM-4.6won 22 · lost 2
Command Awon 21 · lost 2
O3 Miniwon 21 · lost 2
Gemini 2.5 Flash Litewon 17 · lost 2
Mistral Largewon 16 · lost 2
Gemini 3.5 Flash Litewon 16 · lost 2
Claude Haiku 4.5won 23 · lost 3
Inkling Smallwon 14 · lost 2
Grok 4.3won 21 · lost 3
Grok 4.1 Fastwon 19 · lost 3
Gemma 4 31Bwon 18 · lost 3
GPT-5 Miniwon 22 · lost 4
GPT-5.4 Miniwon 16 · lost 3
O4 Miniwon 21 · lost 4
Kimi K2won 21 · lost 4
Qwen3 235B A22B Instructwon 20 · lost 4
GPT-4owon 18 · lost 4
Claude 3.5 Haikuwon 16 · lost 4
Gemini 2.5 Flashwon 20 · lost 5
GPT-4.1won 20 · lost 5
DeepSeek R1won 20 · lost 5
O3won 19 · lost 5
Mimo V2.5 Prowon 18 · lost 5
Claude 3.7 Sonnetwon 17 · lost 5
Claude Sonnet 4won 20 · lost 6
Gemini 2.5 Prowon 20 · lost 6
O1won 16 · lost 5
GLM-4.7won 12 · lost 4
Claude 3.5 Sonnetwon 15 · lost 5
Grok 4won 18 · lost 6
Claude Sonnet 4.5won 19 · lost 7
GLM-5won 13 · lost 5
Kimi K2.5won 18 · lost 7
MiniMax M2.7won 15 · lost 6
MiniMax M3won 17 · lost 7
GPT-5 Nanowon 16 · lost 7
Kimi K2.6won 17 · lost 8
GPT-6 Lunawon 14 · lost 7
Claude Opus 4won 16 · lost 8
GLM-5.1won 14 · lost 8
DeepSeek V4 Flashwon 15 · lost 9
GPT-5.6 Lunawon 15 · lost 9
Qwen3.7 Maxwon 15 · lost 9
Gemini 3.6 Flashwon 15 · lost 9
Qwen3.6 Pluswon 11 · lost 7
Inklingwon 14 · lost 10
GPT-5.1won 14 · lost 10
GLM-5.3 Flashwon 14 · lost 10
GLM-5.2won 14 · lost 10
Gemini 3.5 Flashwon 14 · lost 10
Claude Sonnet 4.6won 13 · lost 10
Qwen3.8 27Bwon 10 · lost 8
GPT-5.2won 13 · lost 11
Gemini 3.8 Flashwon 13 · lost 11
DeepSeek V4 Prowon 12 · lost 13
GPT-5.6 Terrawon 11 · lost 13
GPT-6 Solwon 10 · lost 12
Gemini 3.1 Prowon 9 · lost 11
Claude Opus 4.6won 9 · lost 12
Claude Opus 4.5won 8 · lost 11
GPT-5.4won 10 · lost 15
Gemini 3 Flashwon 6 · lost 10
Grok 4.5won 9 · lost 15
Qwen3.8 Maxwon 9 · lost 15
Mimo V2.6 Prowon 7 · lost 12
Gemini 3 Prowon 8 · lost 14
GPT-5.5won 9 · lost 16
GPT-5.6 Solwon 9 · lost 16
DeepSeek V4.1 Flashwon 6 · lost 11
GLM-5.3won 6 · lost 12
Grok 4.7won 7 · lost 14
Muse Spark 1.1won 7 · lost 17
Grok 4.6won 6 · lost 18
Claude Sonnet 5won 6 · lost 18
Gemini 3.7 Flashwon 4 · lost 14
Claude Opus 4.8won 5 · lost 19
Claude Opus 5won 5 · lost 19
Claude Opus 4.7won 5 · lost 20
Claude Opus 5.5won 4 · lost 17
GPT-6 Astrawon 3 · lost 14
Muse Spark 1.2won 3 · lost 14
Kimi K3won 1 · lost 17
Claude Fable 5.1won 0 · lost 18
Claude Fable 5won 0 · lost 18

§ 3 · Sources

Where the numbers come from

10 publications, 31 figures. Every one links to the page it was read from.

LMArena Text ↗1 measureread 2026-09-26
LMArena Text 1435
GDPval-AA ↗1 measureread 2026-09-26
GDPval-AA 906
ARC-AGI-2 ↗1 measureread 2026-09-26
ARC-AGI-2 9.9%
Aider polyglot ↗1 measureread 2026-09-26
Aider polyglot 88%
Kagi LLM Benchmark ↗1 measureread 2026-09-26
Kagi LLM Benchmark 72.7%
SimpleBench ↗1 measureread 2026-09-26
SimpleBench 56.7%
Vals.ai ↗13 measuresread 2026-09-26
Vals · CaseLaw 66.45%Vals · LegalBench 86.02%Vals · CorpFin 61.07%Vals · TaxEval 73.39%Vals · MortgageTax 65.45%Vals · MedQA 96.32%Vals · MedCode 49.63%Vals · MedScribe 83.65%Vals · SWE-bench Verified 69% (Mini-SWE-agent)Vals · Vibe Code Bench 20.09% (OpenHands)Vals · GPQA Diamond 85.61%Vals · MMLU Pro 86.54%Vals · AIME 93.37%
HELM Safety ↗5 measuresread 2026-09-26
HELM Safety · HarmBench 97.6%HELM Safety · SimpleSafetyTests 99.8%HELM Safety · Anthropic Red Team 99.1%HELM Safety · BBQ 96.8%HELM Safety · XSTest 97.1%
Enkrypt AI Safety Leaderboard ↗6 measuresread 2026-09-26
Enkrypt · Jailbreak risk 2.1%Enkrypt · Harmful content risk 6.7%Enkrypt · CBRN risk 7.1%Enkrypt · Toxicity risk 2.9%Enkrypt · Bias risk 40.1%Enkrypt · Insecure code risk 2.2%
Vectara · Factual consistency 84.9%

Badge

PublicAI Index badge for GPT-5[![PublicAI Index](https://publicai.io/model-index/badge?model=gpt-5)](https://publicai.io/model-index/m/gpt-5)