GPT-5.2

GPT-5.2 ranks first on 11 of the 189 benchmarks it has scores on.

OpenAI Released 11 Dec 2025 Family GPT-5 The lab's model card

194

Benchmarks with a score

Ranked first, out of five or more models

In the top three

OpenAI Dec 2025

GPT-5.2

Closed weights Flagship

153.8 #47 of 271

Epoch Capabilities Index

Humanity's Last Exam 27.8% #14 /47
SWE-bench Verified 73.8% #17 /32
SWE-Bench Pro 29.9% #16 /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
MMLU-Pro	87.4%	#11 of 212	Official leaderboard
Wer wird Millionär? LLM Benchmark	333333.0	#7 of 76	Official leaderboard
LOCA-Bench	21.3%	#1 of 9	Official leaderboard
EQ-Bench 3	1,544	#12 of 78	Official leaderboard
SuperGPQA	61.3%	#9 of 58	Official leaderboard
Judgemark v2.1	82.0	#10 of 42	Official leaderboard
Phare	71%	#15 of 62	Official leaderboard
LMArena (Elo)	1,437	#81 of 333	Official leaderboard
Creative Writing v3	74.0	#35 of 120	Official leaderboard
Humanity's Last Exam	27.8%	#14 of 47	Official leaderboard
Longform Creative Writing (EQ-Bench)	71.6	#37 of 119	Official leaderboard
Extended NYT Connections	83.6%	#34 of 97	Official leaderboard
MultiNRC	42.2%	#15 of 38	Official leaderboard
Vectara Hallucination Leaderboard (2025 dataset)	8.40%	#43 of 104	Official leaderboard
FACTS Search	72.2%	#13 of 31	Official leaderboard
SkateBench v2	69%	#9 of 21	Official leaderboard
OpenAI MRCR v2	77%	#4 of 9	Model card
BullshitBench	27.3%	#51 of 107	Official leaderboard
BullshitBench V2	38%	#60 of 122	Official leaderboard
CL-bench	18.2%	#12 of 24	Official leaderboard
Hallucinations on production-representative prompts	3.10%	#1 of 2	Model card
FACTS Multimodal	39.7%	#21 of 39	Official leaderboard
SimpleQA Verified	37.1%	#52 of 84	Epoch AI
FACTS Benchmark Suite	39.7%	#28 of 40	Official leaderboard
Document Arena	1,425	#34 of 37	Official leaderboard

Math and science

Benchmark	Score	Rank	Source
HMMT 2025	100%	#1 of 53	Official leaderboard
AIME 2025	100%	#6 of 143	Model card
HMMT Nov 2025	99.2%	#1 of 22	Official leaderboard
HiPhO	74.4%	#2 of 35	Official leaderboard
CMIMC 2025	95.6%	#2 of 33	Official leaderboard
HMMT 2026	97%	#3 of 31	Official leaderboard
FrontierMath	40.7%	#7 of 70	Epoch AI
SMT 2025	92%	#4 of 40	Official leaderboard
AIME 2026	98.3%	#4 of 31	Official leaderboard
GPQA Diamond	91.4%	#30 of 219	Epoch AI
BRUMO 2025	98.3%	#6 of 42	Official leaderboard
FrontierMath Tier 4	18.8%	#10 of 55	Epoch AI
OTIS Mock AIME 2024-2025	96.1%	#37 of 202	Epoch AI
ARC-AGI	94.5%	#18 of 92	Official leaderboard
LABBench2	60.9%	#1 of 5	Official leaderboard
DTBench	84.9%	#39 of 178	Official leaderboard
ARC-AGI-2	72.9%	#23 of 95	Official leaderboard
FrontierMath Tiers 1-3 v2	67.4%	#32 of 89	Epoch AI
MathArena (Apex, IMO, USAMO)	13.5%	#18 of 47	Official leaderboard
IMProofBench	32.4%	#4 of 10	Official leaderboard
Echo General AI Prediction Leaderboard	1,001	#8 of 17	Official leaderboard
FrontierMath Tier 4 v2	31.7%	#34 of 71	Epoch AI
IMO-ProofBench	35.7%	#5 of 10	Official leaderboard
SimpleBench	45.8%	#60 of 100	Official leaderboard
ProofBench	15%	#37 of 59	Official leaderboard
ForecastBench	64.0	#60 of 86	Official leaderboard

Coding

Benchmark	Score	Rank	Source
AlgoTune	2.05×	#1 of 18	Official leaderboard
Vibe Code Bench v1	41.3%	#1 of 17	Official leaderboard
OpenAI PRs	55%	#1 of 12	Model card
METR Time Horizon	5.9 h	#4 of 39	Epoch AI
OTelBench	26.1%	#2 of 14	Official leaderboard
IOI (Vals)	54.8%	#8 of 55	Official leaderboard
METR Time Horizon 1.1	5.9 h	#4 of 26	Official leaderboard
WeirdML	72.2%	#18 of 112	Official leaderboard
SWE-fficiency	14.8%	#4 of 17	Official leaderboard
ALE-Bench	1,294	#35 of 121	Epoch AI
OpenHands Index	58.8%	#10 of 33	Official leaderboard
GSO-Bench	27.4%	#11 of 30	Official leaderboard
PostTrainBench	21.4%	#9 of 23	Official leaderboard
WebDev Arena	1,480	#47 of 105	Epoch AI
PaperBench	39%	#7 of 14	Model card
Vibe Code Bench	53.5%	#58 of 115	Official leaderboard
SWE-bench Verified	73.8%	#17 of 32	Epoch AI
MLE-bench	16%	#10 of 18	Model card
TasteVal	27.7×	#12 of 20	Official leaderboard
Code Arena	1,416	#76 of 124	Official leaderboard
FrontierCS	32.4	#9 of 14	Official leaderboard
SWE-Bench Pro	29.9%	#16 of 24	Official leaderboard
OpenAI-Proof Q&A	3.00%	#6 of 9	Model card
SWE-bench Multilingual	66.7%	#9 of 13	Official leaderboard
Vals RSI Index	0.16	#22 of 25	Official leaderboard
WeirdML v3	5.01%	#23 of 25	Official leaderboard

Agents and tools

Benchmark	Score	Rank	Source
Elimination Game	8	#1 of 56	Official leaderboard
MCPMark	57.5%	#1 of 28	Official leaderboard
Agent Poker Bench	1,132	#1 of 17	Official leaderboard
Pencil Puzzle Bench	56%	#4 of 53	Official leaderboard
GAIA	86%	#5 of 57	Official leaderboard
SnakeBench	33	#21 of 183	Official leaderboard
BalatroBench	18.4	#2 of 16	Official leaderboard
RWSearch	82%	#1 of 7	Official leaderboard
ScreenSpot-Pro	86.3%	#4 of 25	Model card
GameWorld	40.6%	#2 of 12	Official leaderboard
LLM Chess	1,085	#22 of 124	Official leaderboard
Terminal-Bench 2.0	64.9%	#8 of 43	Official leaderboard
BFCL V4	55.9%	#13 of 65	Official leaderboard
tau2-bench	90.3%	#14 of 69	Model card
Chess Puzzles	48%	#18 of 83	Epoch AI
LisanBench	3592.3	#32 of 133	Official leaderboard
Game Arena Five in a Row	817	#5 of 20	Official leaderboard
Toolathlon	43.8%	#14 of 50	Official leaderboard
Game Arena Poker	4.98	#9 of 31	Official leaderboard
Game Arena Clobber	348	#12 of 36	Official leaderboard
VITA-Bench	24.3%	#6 of 18	Official leaderboard
Game Arena Chess Openings	889	#4 of 11	Official leaderboard
Game Arena Checkers	967	#15 of 38	Official leaderboard
Game Arena Reversi	616	#15 of 36	Official leaderboard
SkillsBench	51.7%	#9 of 21	Official leaderboard
tau3-bench	84.8%	#3 of 7	Official leaderboard
Game Arena Chess	825	#17 of 39	Official leaderboard
Game Arena Lines of Action	543	#16 of 36	Official leaderboard
Game Arena Four in a Row	393	#17 of 38	Official leaderboard
Mystery Game Puzzles	23%	#40 of 80	Epoch AI
Kaggle Game Arena	283	#20 of 39	Official leaderboard
τ³-bench Banking	32.2%	#13 of 25	Official leaderboard
Game Arena Dark Hex	247	#20 of 38	Official leaderboard
Game Arena Dots and Boxes	879	#20 of 37	Official leaderboard
Game Arena Word Association	189	#21 of 36	Official leaderboard
EBR-bench	23%	#17 of 28	Epoch AI
Game Arena Ultimate Tic-Tac-Toe	244	#22 of 36	Official leaderboard
Game Arena Go	557	#24 of 36	Official leaderboard
MCP Atlas	67.6%	#24 of 34	Official leaderboard
Game Arena Coin Game	241	#26 of 35	Official leaderboard
Game Arena Word Art	248	#26 of 35	Official leaderboard
Game Arena Nine Men's Morris	510	#27 of 36	Official leaderboard
DeepResearch Bench	41.1%	#20 of 24	Epoch AI
LMArena Search Arena	1,161	#31 of 35	Official leaderboard
Game Arena Bargaining	131	#32 of 36	Official leaderboard

Professional work

Benchmark	Score	Rank	Source
TaxEval v2	75.8%	#11 of 129	Official leaderboard
Vals Finance Agent v1	52.8%	#5 of 47	Official leaderboard
CaseLaw v2	66%	#8 of 53	Official leaderboard
HealthBench	63.3%	#5 of 30	Model card
Vals Finance Agent	58.5%	#9 of 49	Official leaderboard
MedCode	49.8%	#20 of 96	Official leaderboard
CorpFin v2	65.9%	#26 of 119	Official leaderboard
HealthBench Hard	42%	#5 of 21	Model card
ProfBench	55.1%	#18 of 72	Official leaderboard
SpreadsheetBench 2	26.8%	#2 of 8	Official leaderboard
prinzbench	52.0	#7 of 26	Official leaderboard
MortgageTax	67.1%	#28 of 86	Official leaderboard
MedScribe	84.4%	#35 of 98	Official leaderboard
LegalBench	82.8%	#65 of 138	Official leaderboard
TaxCalcBench	33.8%	#10 of 17	Official leaderboard
Vending-Bench 2	$3,591	#46 of 66	Official leaderboard
GDPval	70.9%	#7 of 10	Model card
Remote Labor Index	2.50%	#12 of 17	Official leaderboard
HealthBench Consensus	94.5%	#14 of 17	Model card
Vending-Bench Arena	$619	#29 of 32	Official leaderboard

Vision, audio and robotics

Benchmark	Score	Rank	Source
MMMU-Pro	80.4%	#4 of 57	Official leaderboard
VPCT	84%	#2 of 28	Epoch AI
MATH-Vision (MathVision)	83%	#16 of 105	Official leaderboard
CharXiv	88.7%	#11 of 67	Model card
ZeroBench	18%	#15 of 66	Official leaderboard
ERQA	41.5%	#4 of 17	Official leaderboard
OCRBench v2	50.5%	#17 of 65	Official leaderboard
RoboRewardBench	0.89	#5 of 19	Official leaderboard
Math Kangaroo 2025 (MathArena)	86.5%	#7 of 22	Official leaderboard
VideoDR (with search)	69%	#2 of 6	Official leaderboard
EnigmaEval	10.4%	#14 of 41	Epoch AI
Vision Arena	1,243	#52 of 136	Official leaderboard
VISTA	46.6%	#21 of 54	Official leaderboard
MADQA	67.8%	#7 of 16	Official leaderboard
Furniture Assembly Benchmark	38.3%	#19 of 35	Epoch AI
ClockBench	15.6%	#21 of 37	Official leaderboard
BabyVision	34.4%	#17 of 27	Official leaderboard
MDPBench	68.6	#6 of 9	Official leaderboard
Drone-Bench	51.7%	#13 of 19	Official leaderboard
OmniDocBench v1.6	86.6	#5 of 7	Official leaderboard
VoxelBench	1,426	#42 of 56	Official leaderboard
OmniDocBench v1.5	85.8	#8 of 10	Official leaderboard
OCR Arena	1,359	#11 of 11	Official leaderboard

Safety and security

Benchmark	Score	Rank	Source
Spiral-Bench v1.2	70.2	#3 of 21	Official leaderboard
StrongREJECT	97.5%	#4 of 22	Model card
KORA	61.4%	#9 of 42	Official leaderboard
ProtocolQA Open-Ended	40%	#7 of 25	Model card
Capture the Flag (Professional)	82%	#6 of 21	Model card
First-Person Fairness evaluation	1.00%	#3 of 10	Model card
FORTRESS	17.5%	#20 of 66	Official leaderboard
Human Pathogen Capabilities Test v2 (HPCT v2)	63.2%	#22 of 71	Official leaderboard
Deception in production traffic	1.60%	#1 of 2	Model card
PlugInject	99.6%	#1 of 2	Model card
TroubleshootingBench	35%	#10 of 19	Model card
BinaryAudit	17.5%	#13 of 20	Official leaderboard
CVE-Bench	69%	#11 of 14	Model card
DrugDiscoveryBench	29.3%	#16 of 19	Official leaderboard
Agent JSK	97.8%	#6 of 7	Model card
Broken Search Tool	9.10%	#6 of 7	Model card
Multimodal Troubleshooting Virology	43%	#23 of 26	Model card
PCT (Polarized Contrastive Pairs)	57%	#41 of 46	Official leaderboard
Tacit Knowledge and Troubleshooting	48%	#29 of 29	Model card
Coding Deception	25.6%	#6 of 6	Model card
CharXiv Missing Image	88.8%	#2 of 2	Model card
Production Deception - Adversarial	5.40%	#1 of 1	Model card

Other

Benchmark	Score	Rank	Source
ACCoRD	79.3	#14 of 150	Official leaderboard
Conceptual Reasoning Index	59.2	#39 of 149	Official leaderboard
SAGE (Vals)	49.3%	#22 of 81	Official leaderboard
LMCA	43.9%	#43 of 139	Epoch AI
CAIS Text Capabilities Index	33.5	#30 of 66	Official leaderboard
CAIS Risk Index	50.3	#30 of 65	Official leaderboard
CAIS Vision Capabilities Index	52.2	#26 of 53	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-5.2 11 Dec 2025

Its system card

Every result in the longer documents the lab published with it. 29 of them aren't in the launch post.

GPT-5.2

GPT-5.2

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 card

Capabilities

Health

Hallucination

Safety

Robustness

Bias

Alignment

Bio and chem risk

Cyber risk

AI R&D risk