GPT-5

GPT-5 ranks first on 23 of the 201 benchmarks it has scores on.

OpenAI Released 7 Aug 2025 Family GPT-5 The lab's model card

223

Benchmarks with a score

Ranked first, out of five or more models

In the top three

OpenAI Aug 2025

GPT-5

Closed weights Flagship

150.0 #63 of 271

Epoch Capabilities Index

Humanity's Last Exam 25.3% #15 /47
SWE-bench Verified 73.5% #19 /32
SWE-Bench Pro 41.8% #10 /24

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

Benchmark	Score	Rank	Source
LLM Confabulation Leaderboard	10.3%	#1 of 58	Official leaderboard
WritingBench	83.9	#1 of 45	Official leaderboard
Lech Mazur Creative Writing	8.60	#1 of 44	Epoch AI
Fiction.LiveBench	97.2%	#2 of 57	Epoch AI
MMBench-CN v1.1	86.4%	#8 of 178	Official leaderboard
FACTS Grounding	87.4%	#2 of 34	Official leaderboard
Wer wird Millionär? LLM Benchmark	366667.0	#5 of 76	Official leaderboard
SuperGPQA	65.6%	#4 of 58	Official leaderboard
MMLU-Pro	87.1%	#16 of 212	Official leaderboard
COLLIE	99%	#1 of 12	Model card
SimpleQA	55%	#10 of 59	Model card
SkateBench v2	77.4%	#4 of 21	Official leaderboard
FACTS Search	77.7%	#6 of 31	Official leaderboard
FACTS Multimodal	44.1%	#9 of 39	Official leaderboard
MultiNRC	52.1%	#9 of 38	Official leaderboard
FActScore	1.00%	#1 of 4	Model card
LongFact-Objects	1.40%	#1 of 4	Model card
LMArena (Elo)	1,435	#84 of 333	Official leaderboard
Longform Creative Writing (EQ-Bench)	72.8	#34 of 119	Official leaderboard
Creative Writing v3	72.7	#38 of 120	Official leaderboard
Humanity's Last Exam	25.3%	#15 of 47	Official leaderboard
SimpleQA Verified	50.1%	#28 of 84	Epoch AI
MultiChallenge	63.2%	#13 of 35	Official leaderboard
Vectara Hallucination Leaderboard (HHEM)	4.90%	#58 of 148	Official leaderboard
FACTS Parametric	55.8%	#18 of 45	Official leaderboard
LongFact	0.80%	#2 of 5	Model card
SimpleQA Hallucination Rate	40%	#5 of 12	Model card
Phare	67.2%	#28 of 62	Official leaderboard
IndicGenBench	52.2	#6 of 13	Official leaderboard
HealthBench Hard Hallucinations	1.60%	#1 of 2	Model card
FACTS Benchmark Suite	44.1%	#21 of 40	Official leaderboard
Global MMLU Lite	89%	#23 of 43	Official leaderboard
BullshitBench	21.8%	#66 of 107	Official leaderboard
BullshitBench V2	21%	#83 of 122	Official leaderboard
LongFact-Concepts	1.10%	#3 of 4	Model card
Vectara Hallucination Leaderboard (2025 dataset)	14.7%	#94 of 104	Official leaderboard
AbstentionBench	53%	#2 of 2	Model card
Hallucinations on production-representative prompts	4.50%	#2 of 2	Model card

Math and science

Benchmark	Score	Rank	Source
MATH Level 5	98.1%	#1 of 94	Epoch AI
AIME 2025	100%	#2 of 143	Model card
HMMT 2025	100%	#1 of 49	Model card
AMO-Bench	52.4%	#1 of 21	Official leaderboard
SMT 2025	92%	#3 of 40	Official leaderboard
HiPhO	67.9%	#4 of 35	Official leaderboard
PHYBench	40.4%	#4 of 28	Official leaderboard
IMO 2025	38.1%	#1 of 6	Official leaderboard
General365	58.6%	#4 of 23	Official leaderboard
ATLAS (OpenCompass)	42.9%	#1 of 5	Official leaderboard
FrontierMath	32.4%	#16 of 70	Epoch AI
DTBench	84.5%	#42 of 178	Official leaderboard
OTIS Mock AIME 2024-2025	91.4%	#55 of 202	Epoch AI
GPQA Diamond	86.2%	#66 of 219	Epoch AI
CMIMC 2025	88.8%	#10 of 33	Official leaderboard
FrontierMath Tier 4	12.5%	#17 of 55	Epoch AI
CritPt	12.6%	#52 of 160	Epoch AI
SimpleBench	56.7%	#44 of 100	Official leaderboard
BRUMO 2025	91.7%	#20 of 42	Official leaderboard
CORE-Bench	26.7%	#9 of 17	Official leaderboard
ProofBench	18%	#33 of 59	Official leaderboard
FrontierMath Tiers 1-3 v2	55.4%	#50 of 89	Epoch AI
ARC-AGI	65.7%	#53 of 92	Official leaderboard
HMMT Nov 2025	89.2%	#13 of 22	Official leaderboard
ARC-AGI-2	9.86%	#57 of 95	Official leaderboard
IMProofBench	20.4%	#6 of 10	Official leaderboard
MathArena (Apex, IMO, USAMO)	2.08%	#31 of 47	Official leaderboard
FrontierMath Tier 4 v2	21.9%	#47 of 71	Epoch AI
ForecastBench	65.2	#75 of 86	Official leaderboard
CritPt v2	28.9%	#11 of 11	Official leaderboard

Coding

Benchmark	Score	Rank	Source
Aider Polyglot	88%	#1 of 54	Epoch AI
ArtifactsBench	72.5%	#1 of 20	Official leaderboard
METR Time Horizon	2.2 h	#3 of 31	Official leaderboard
SWE-fficiency	15.7%	#2 of 17	Official leaderboard
CooperBench	27.9%	#1 of 6	Official leaderboard
CodeClash	1,366	#2 of 8	Official leaderboard
WeirdML	60.7%	#32 of 112	Official leaderboard
SWE-Lancer	55%	#3 of 10	Model card
OpenAI PRs	45%	#4 of 12	Model card
ALE-Bench	1,162	#45 of 121	Epoch AI
METR Time Horizon 1.1	3.4 h	#10 of 26	Official leaderboard
SWE-Bench Pro	41.8%	#10 of 24	Official leaderboard
IOI (Vals)	20%	#23 of 55	Official leaderboard
LiveSQLBench	25.5%	#16 of 36	Official leaderboard
AlgoTune	1.67×	#8 of 18	Official leaderboard
Vibe Code Bench v1	20.1%	#8 of 17	Official leaderboard
SWE-bench Verified	73.5%	#19 of 32	Epoch AI
WebDev Arena	1,420	#63 of 105	Epoch AI
Code Arena	1,417	#75 of 124	Official leaderboard
GSO-Bench	6.86%	#20 of 30	Official leaderboard
SWE-bench-Live	24.3%	#9 of 13	Official leaderboard
PaperBench	24%	#10 of 14	Model card
Vibe Code Bench	20.1%	#83 of 115	Official leaderboard
TasteVal	11.6×	#15 of 20	Official leaderboard
OpenAI-Proof Q&A	2.00%	#7 of 9	Model card
FrontierCS	23.1	#12 of 14	Official leaderboard
MLE-bench	8.00%	#17 of 18	Model card
MirrorCode	4.44%	#28 of 28	Epoch AI

Agents and tools

Benchmark	Score	Rank	Source
Step Race	5	#1 of 59	Official leaderboard
Elimination Game	6	#2 of 56	Official leaderboard
SnakeBench	35	#9 of 183	Official leaderboard
Terminal-Bench	49.6%	#1 of 18	Epoch AI
MCP-Universe	44.2%	#2 of 30	Official leaderboard
Meta Agents Research Environments (ARE) / Gaia2	42.1%	#1 of 13	Official leaderboard
MCPMark	52.6%	#3 of 28	Official leaderboard
Agent Poker Bench	1,103	#2 of 17	Official leaderboard
LiveResearchBench	73.7	#1 of 7	Official leaderboard
CompileBench	93.3%	#3 of 20	Official leaderboard
GAIA	78.4%	#9 of 57	Official leaderboard
LLM Chess	1,086	#21 of 124	Official leaderboard
VitaBench 2.0	44.1%	#4 of 20	Official leaderboard
ITBench	51.9%	#4 of 13	Official leaderboard
LisanBench	2247.0	#41 of 133	Official leaderboard
DeepResearch Bench	49.6%	#8 of 24	Epoch AI
AgentBench FC	52.2%	#4 of 12	Official leaderboard
OSWorld-Verified	69.9%	#12 of 33	Official leaderboard
BALROG	32.8%	#14 of 37	Epoch AI
Terminal-Bench 2.0	49.6%	#17 of 43	Official leaderboard
DeepSearchQA	59.4%	#4 of 10	Official leaderboard
Pencil Puzzle Bench	3.30%	#24 of 53	Official leaderboard
Toolathlon	37.7%	#24 of 50	Official leaderboard
Mystery Game Puzzles	23%	#39 of 80	Epoch AI
tau2-bench	80.1%	#34 of 69	Model card
MobileWorld	54%	#9 of 18	Official leaderboard
Chess Puzzles	36%	#42 of 83	Epoch AI
ScreenSpot-Pro	18.5%	#7 of 11	Official leaderboard
BrowseComp	54.9%	#56 of 77	Model card
LMArena Search Arena	1,188	#26 of 35	Official leaderboard
EBR-bench	12.7%	#21 of 28	Epoch AI

Professional work

Benchmark	Score	Rank	Source
LEXam	70.2%	#1 of 31	Official leaderboard
HealthBench	67.2%	#1 of 30	Model card
HealthBench Hard	46.2%	#1 of 21	Model card
MedHELM	70.3	#1 of 13	Official leaderboard
CaseLaw v2	66.5%	#6 of 53	Official leaderboard
LegalBench	86%	#18 of 138	Official leaderboard
Vals Finance Agent v1	46.9%	#9 of 47	Official leaderboard
MedCode	49.6%	#21 of 96	Official leaderboard
FailureSensorIQ	67.2%	#6 of 24	Official leaderboard
TaxCalcBench	41.7%	#5 of 17	Official leaderboard
PRBench	49%	#14 of 44	Official leaderboard
Vending-Bench	$3,579	#3 of 9	Official leaderboard
HealthBench Consensus	95.7%	#6 of 17	Model card
TaxEval v2	73.4%	#47 of 129	Official leaderboard
ProfBench	49.4%	#30 of 72	Official leaderboard
MedScribe	83.7%	#42 of 98	Official leaderboard
MortgageTax	65.5%	#39 of 86	Official leaderboard
HealthBench Consensus Global Health	0%	#1 of 2	Model card
HealthBench Consensus Urgent	0.40%	#1 of 2	Model card
CorpFin v2	61.1%	#60 of 119	Official leaderboard
Vals Finance Agent	52.1%	#25 of 49	Official leaderboard
GDPval	34.8%	#6 of 11	Epoch AI
StockBench	0.30%	#9 of 13	Official leaderboard
Remote Labor Index	1.67%	#14 of 17	Official leaderboard

Vision, audio and robotics

Benchmark	Score	Rank	Source
MMStar	75.7%	#1 of 173	Official leaderboard
MMMU	84.2%	#2 of 136	Official leaderboard
MMBench-EN v1.1	86.8%	#9 of 178	Official leaderboard
Blueprint-Bench	0.43	#1 of 8	Official leaderboard
VPCT	66%	#4 of 28	Epoch AI
RoboRewardBench	0.81	#3 of 19	Official leaderboard
MADQA	77.7%	#3 of 16	Official leaderboard
OCRBench v2	55.5%	#14 of 65	Official leaderboard
MATH-Vision (MathVision)	72%	#23 of 105	Official leaderboard
VISTA	49.7%	#12 of 54	Official leaderboard
VideoMME v2 (with subtitles)	37%	#7 of 31	Official leaderboard
MMMU-Pro	78.4%	#19 of 72	Model card
Math-VR	58.1%	#7 of 26	Official leaderboard
GeoBench	81%	#13 of 48	Official leaderboard
EnigmaEval	10.5%	#13 of 41	Epoch AI
Video-MMMU	84.6%	#10 of 30	Model card
ERQA	65.7%	#10 of 26	Model card
CharXiv	81.1%	#27 of 67	Model card
ZeroBench	7.00%	#33 of 66	Official leaderboard
Vision Arena	1,211	#72 of 136	Official leaderboard
Math Kangaroo 2025 (MathArena)	78.8%	#12 of 22	Official leaderboard
ClockBench	11.1%	#26 of 37	Official leaderboard
OCR Arena	1,382	#8 of 11	Official leaderboard
Drone-Bench	37.8%	#15 of 19	Official leaderboard
VoxelBench	1,307	#49 of 56	Official leaderboard

Safety and security

Benchmark	Score	Rank	Source
HELM Safety	98.1	#5 of 75	Official leaderboard
Spiral-Bench v1.0	87.0	#1 of 13	Official leaderboard
AIR-Bench 2024	87.7%	#7 of 82	Official leaderboard
BaxBench	54.3%	#3 of 34	Official leaderboard
PCT (Polarized Contrastive Pairs)	35.3%	#7 of 46	Official leaderboard
Spiral-Bench v1.2	57.4	#6 of 21	Official leaderboard
FORTRESS	17%	#19 of 66	Official leaderboard
SimpleSafetyTests	99.8%	#25 of 75	Official leaderboard
Capture the Flag (Collegiate)	31%	#6 of 13	Model card
Anthropic Red Team	99.1%	#36 of 75	Official leaderboard
CyberGym	60.2%	#16 of 33	Official leaderboard
Apollo Research strategic deception evaluations	3.97%	#1 of 2	Model card
CharXiv Missing Image	9.00%	#1 of 2	Model card
Capture the Flag (Professional)	27%	#12 of 21	Model card
Challenging prompts from red teamers with biosafety-relevant PhDs	92.1%	#2 of 3	Model card
ProtocolQA Open-Ended	30%	#17 of 25	Model card
Virology Capabilities Test (VCT)	41.4%	#23 of 28	Model card
Coding Deception	17%	#5 of 6	Model card
Molecular Biology Capabilities Test	48%	#6 of 7	Model card
Tacit Knowledge and Troubleshooting	66%	#25 of 29	Model card
Human Pathogen Capabilities Test	53.7%	#7 of 8	Model card
Multimodal Troubleshooting Virology	42%	#24 of 26	Model card
TroubleshootingBench	22%	#18 of 19	Model card
BinaryAudit	1.75%	#20 of 20	Official leaderboard
Gray Swan Agent Red Teaming (ART) benchmark	56.8%	#14 of 14	Model card
Cyber Range	0%	#10 of 10	Model card
Broken Search Tool	11%	#7 of 7	Model card
Coding prompt injections	97%	#4 of 4	Model card
Filtered, adversarial sample of production prompts	95.7%	#3 of 3	Model card
World-Class Bio	47.6%	#3 of 3	Model card
Deception in production traffic	2.10%	#2 of 2	Model card
Browsing prompt injections	99%	#1 of 1	Model card
Expert Red Teaming for Violent Attack Planning	65.1%	#1 of 1	Model card
Pairwise Bioweaponization Campaign	60%	#1 of 1	Model card
Sycophancy evaluation (offline)	0.04	#1 of 1	Model card
Tool-calling prompt injections	99%	#1 of 1	Model card
Virology Capabilities Test (text-only)	46%	#1 of 1	Model card

Other

Benchmark	Score	Rank	Source
OpenVLM Leaderboard	79.9	#2 of 173	Official leaderboard
HELM Capabilities	80.7	#4 of 60	Official leaderboard
ACCoRD	73.1	#33 of 150	Official leaderboard
Conceptual Reasoning Index	55.5	#50 of 149	Official leaderboard
CAIS Risk Index	44.9	#22 of 65	Official leaderboard
LMCA	40%	#51 of 139	Epoch AI
AGI Score (A Definition of AGI)	57%	#1 of 2	Official leaderboard
CAIS Text Capabilities Index	27.7	#36 of 66	Official leaderboard
CAIS Vision Capabilities Index	47.8	#30 of 53	Official leaderboard
SAGE (Vals)	43.7%	#47 of 81	Official leaderboard

OpenAI's models over time

OpenAI's models in color, against every other model on the Epoch Capabilities Index, by release date.

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.

Introducing GPT-5 7 Aug 2025

Its system cards

Every result in the longer documents the lab published with it. 50 of them aren't in the launch post.

GPT-5 System Card 7 Aug 2025

Capabilities

MMMLU

Health

Hallucination

Hallucinations on production-representative prompts 4.5
LongFact-Concepts 1.1
LongFact-Objects 1.4
FActScore 3.7
SimpleQA 55
SimpleQA Hallucination Rate 40

GPT-5

GPT-5

Where it ranks

Knowledge and language

Math and science

Coding

Agents and tools

Professional work

Vision, audio and robotics

Safety and security

Other

OpenAI's models over time

Its launch post

Its system cards

Capabilities

Health

Hallucination

Safety

Robustness

Bias

Alignment

Bio and chem risk

Cyber risk

AI R&D risk

Autonomy

Hallucination

Safety

Robustness