benchgap
OpenAI · model

GPT-5.2 benchmark scores

As of 2026-11-21, GPT-5.2 (OpenAI) has measured scores on 57 benchmarks and estimated scores on 122 more.

BenchmarkScoreSource
AA-GPQA Diamond90.3%measured
AA-HLE37.7%measured
AA-SciCode48.0%estimated ± 3.6 pp, high confidence
AIME2699.2%estimated ± 1.2 pp, low confidence
Artificial Analysis Intelligence Index30.5%measured
CritPt11.6%measured
GDPval-AA36.8%estimated ± 6.2 pp, medium confidence
GPQA Diamond92.4%estimated ± 0.1 pp, high confidence
HLE44.8%estimated ± 6.1 pp, medium confidence
IFBench73.6%estimated ± 6.6 pp, medium confidence
AA-LCR82.7%measured
LiveCodeBench v688.9%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy44.3%measured
SciCode47.5%estimated ± 2.9 pp, medium confidence
BrowseComp65.8%measured
HLE w/ tools38.4%estimated ± 5.5 pp, medium confidence
IFEval93.0%estimated ± 2.8 pp, high confidence
LongBench v268.1%estimated ± 5.5 pp, medium confidence
VITA-Bench37.2%estimated ± 6.7 pp, low confidence
τ²-bench results84.8%measured
AA Coding Index60.9%estimated ± 5.3 pp, medium confidence
AA-MMMU-Pro78.5%estimated ± 1.0 pp, medium confidence
APEX-Agents-AA31.9%estimated ± 3.6 pp, medium confidence
BioMysteryBench (human-difficult)44.9%estimated ± 6.8 pp, low confidence
DeepSearchQA78.4%estimated ± 6.8 pp, low confidence
SWE-bench Verified80.0%measured
BFCL v463.9%estimated ± 8.5 pp, low confidence
CyberGym52.7%estimated ± 7.3 pp, medium confidence
JobBench34.3%measured
SkillsBench37.7%estimated ± 4.1 pp, low confidence
SWE-bench Pro55.6%measured
WideResearch77.8%estimated ± 1.2 pp, medium confidence
AA Agentic Index14.4%estimated ± 6.8 pp, medium confidence
MMLU-Pro85.2%estimated ± 4.4 pp, high confidence
AA-IFBench75.4%measured
AIME 202496.9%estimated ± 2.6 pp, medium confidence
FailureSensorIQ67.6%estimated ± 1.7 pp, high confidence
GPQA92.4%measured
MMLU90.5%estimated ± 0.9 pp, medium confidence
Vals GPQA Diamond91.7%measured
Vals LiveCodeBench85.4%measured
Vals MMLU-Pro86.2%measured
MATH-50099.4%estimated ± 2.0 pp, medium confidence
MMMLU89.8%estimated ± 0.7 pp, medium confidence
MMMU86.7%measured
Chess Puzzles48.0%measured
AIME 2025100.0%measured
AA MMLU-Pro87.4%measured
Gert Labs46.5%measured
AA EnterpriseOps-Gym44.7%estimated ± 1.2 pp, medium confidence
AA Harvey LAB v1.086.9%estimated ± 9.8 pp, low confidence
AA Tau3 Banking26.2%estimated ± 3.6 pp, low confidence
ApprenticeBench5.0%estimated ± 6.0 pp, low confidence
ARC-AGI-186.2%measured
ARC-AGI-252.9%measured
BabaIsBench95.9%estimated ± 1.2 pp, high confidence
CursorBench 3.151.7%estimated ± 5.6 pp, low confidence
CursorBench 3.251.4%estimated ± 3.6 pp, high confidence
EBR-bench23.0%measured
FrontierCode 1.1 Main27.5%estimated ± 4.2 pp, medium confidence
FrontierMath v2 (Tier 4)31.7%measured
FrontierMath v2 (Tiers 1-3)67.4%measured
FrontierSWE v28.0%estimated ± 8.3 pp, medium confidence
Furniture Assembly38.3%measured
MaxxBench52.7%estimated ± 4.3 pp, low confidence
MazeBench0.0%estimated ± 5.1 pp, medium confidence
MirrorCode32.2%measured
Mystery Game Puzzles23.0%measured
NYT Connections83.6%measured
OSWorld-Verified47.3%measured
ParseBench-163.6%estimated ± 8.4 pp, medium confidence
Terminal-Bench 4.12.2%estimated ± 9.2 pp, low confidence
Terminal-Bench Hard47.0%measured
Vals SWE-bench75.8%measured
Terminal-Bench 2.1 (Vals AI)58.6%estimated ± 5.6 pp, medium confidence
VulcanBench v383.9%estimated ± 5.6 pp, low confidence
CWE-Bench-AA43.6%estimated ± 6.4 pp, medium confidence
AA Cyber Index32.8%estimated ± 13.5 pp, low confidence
DeepsecBench-AA4.5%estimated ± 9.0 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)55.1%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)57.6%estimated ± 0.8 pp, low confidence
AA ITBench37.0%estimated ± 2.8 pp, high confidence
Bug Hunt Bench29.3%estimated ± 12.1 pp, low confidence
CursorBench 4.032.3%estimated ± 3.3 pp, medium confidence
DeepSWE56.3%estimated ± 7.2 pp, low confidence
DRACO65.1%estimated ± 2.9 pp, low confidence
FrontierCode 2.020.0%estimated ± 6.7 pp, low confidence
FrontierPhysics2.4%estimated ± 4.4 pp, high confidence
HLE w/o tools34.4%estimated ± 4.9 pp, high confidence
OSWorld 2.00.0%estimated ± 5.9 pp, low confidence
PostTrainBench v1.121.4%estimated ± 1.3 pp, medium confidence
SpireClimb5.8%estimated ± 7.2 pp, medium confidence
StealthMark46.1%estimated ± 6.5 pp, medium confidence
Toolathlon-Verified50.5%estimated ± 7.4 pp, medium confidence
VulcanBench Frontier v477.1%estimated ± 8.3 pp, low confidence
GDP.pdf16.4%estimated ± 4.1 pp, medium confidence
ARC-AGI-31.3%estimated ± 3.1 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
SkillsBench official leaderboard (BenchFlow)51.7%measured
BinaryAudit17.5%measured
CompileBench91.1%estimated ± 10.0 pp, low confidence
OTelBench26.1%measured
Vibe Code Bench53.5%measured
FACTS Multimodal47.0%estimated ± 4.0 pp, high confidence
ParseBench-271.3%estimated ± 5.0 pp, medium confidence
CharXiv82.1%measured
CharXiv w/o tools83.6%estimated ± 1.2 pp, high confidence
ExploitBench32.4%estimated ± 6.5 pp, medium confidence
AI2D_TEST92.2%measured
C-Eval94.7%estimated ± 1.0 pp, low confidence
Claw-Eval63.3%estimated ± 4.5 pp, medium confidence
CountBench91.9%measured
DynaMath86.8%measured
ERQA59.8%measured
HMMT Feb 202596.6%estimated ± 2.9 pp, medium confidence
HMMT Feb 202695.4%estimated ± 1.5 pp, medium confidence
HMMT Nov 202595.4%estimated ± 1.5 pp, low confidence
MathVision83.0%measured
MCP Atlas63.6%estimated ± 8.6 pp, medium confidence
MLVU (M-Avg)85.6%measured
MMAnswerBench91.0%estimated ± 2.4 pp, low confidence
MMLU-Redux96.5%estimated ± 0.8 pp, low confidence
MMMU-Pro79.5%measured
NL2Repo42.9%estimated ± 4.7 pp, high confidence
OmniDocBench 1.585.7%measured
PostTrainBench v121.4%measured
RealWorldQA83.3%measured
ScreenSpot Pro68.8%estimated ± 4.8 pp, medium confidence
SimpleVQA55.8%measured
SuperGPQA78.6%estimated ± 8.5 pp, low confidence
Terminal-Bench 2.064.8%estimated ± 5.4 pp, medium confidence
Toolathlon36.4%estimated ± 4.2 pp, high confidence
V*75.9%measured
Video-MME (w/o subtitle)85.8%measured
Video-MME (with subtitle)86.0%measured
VideoMMMU85.9%measured
We-Math79.0%measured
AA LiveCodeBench89.3%estimated ± 2.6 pp, high confidence
HealthBench Hard32.1%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro80.2%estimated ± 6.0 pp, low confidence
MedXpertQA (MM)73.5%estimated ± 2.4 pp, high confidence
React Native Evals79.6%estimated ± 2.7 pp, medium confidence
ResearchClawBench11.9%estimated ± 2.1 pp, low confidence
SWE-Rebench61.8%estimated ± 3.6 pp, medium confidence
PostTrainBench18.8%estimated ± 1.2 pp, medium confidence
DAYJOB1.4%estimated ± 2.7 pp, high confidence
DAYJOB Healthcare0.0%estimated ± 4.0 pp, high confidence
BioMysteryBench (human-solvable)84.7%estimated ± 0.6 pp, low confidence
FrontierCode 1.1 Extended54.5%estimated ± 3.0 pp, low confidence
HealthBench (raw)46.6%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)53.6%estimated ± 2.5 pp, low confidence
HealthBench Professional55.8%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)57.3%estimated ± 4.5 pp, medium confidence
HLE-Verified20.0%estimated ± 2.0 pp, low confidence
LABBench281.0%estimated ± 1.8 pp, low confidence
ScarfBench46.7%estimated ± 12.3 pp, low confidence
BlenderWorld (text-only)70.2%estimated ± 2.8 pp, medium confidence
Cookie-Bench77.4%estimated ± 1.3 pp, medium confidence
STEMBench2.5%estimated ± 0.2 pp, medium confidence
SimpleQA51.5%estimated ± 4.3 pp, low confidence
LiveCodeBench79.3%estimated ± 14.6 pp, low confidence
Agents' Last Exam16.3%estimated ± 11.6 pp, low confidence
BabyVision w/ Python88.8%estimated ± 3.8 pp, low confidence
OpenHarmony Bench v1.053.0%estimated ± 3.2 pp, high confidence
SEC-Bench Pro42.4%estimated ± 7.5 pp, low confidence
BabyVision40.1%estimated ± 3.7 pp, high confidence
MMVU78.1%estimated ± 2.7 pp, medium confidence
PerceptionBench49.7%estimated ± 3.6 pp, low confidence
ZeroBench15.7%estimated ± 3.6 pp, low confidence
MRCRv285.9%estimated ± 11.3 pp, low confidence
CC-OCR70.3%measured
MMLongBench-Doc61.5%estimated ± 2.6 pp, low confidence
RefCOCO (avg)90.1%estimated ± 3.1 pp, medium confidence
OCRBench V271.5%estimated ± 2.1 pp, medium confidence
BioBasicsBench31.2%measured
LifeVal49.3%measured
MMMU-Pro w/ Python81.5%estimated ± 2.6 pp, high confidence
ExploitBench Internal Port1.4%estimated ± 1.4 pp, medium confidence
PinchBench89.3%estimated ± 3.6 pp, medium confidence