benchgap
OpenAI · model

GPT-5.4 Pro benchmark scores

As of 2026-11-22, GPT-5.4 Pro (OpenAI) has measured scores on 9 benchmarks and estimated scores on 76 more.

BenchmarkScoreSource
AA-GPQA Diamond93.4%estimated ± 2.7 pp, high confidence
AA-HLE41.7%estimated ± 4.6 pp, high confidence
AIME2697.9%estimated ± 1.4 pp, medium confidence
Artificial Analysis Intelligence Index33.2%estimated ± 5.6 pp, medium confidence
CritPt30.0%measured
GDPval-AA51.8%estimated ± 7.8 pp, medium confidence
GPQA Diamond93.1%estimated ± 1.4 pp, high confidence
HLE58.7%measured
AA-Omniscience Accuracy53.2%estimated ± 7.4 pp, medium confidence
BrowseComp89.3%measured
HLE w/ tools61.5%estimated ± 5.6 pp, medium confidence
VITA-Bench47.0%estimated ± 6.7 pp, low confidence
τ²-bench results93.6%estimated ± 11.8 pp, low confidence
APEX-Agents-AA39.7%estimated ± 4.4 pp, high confidence
BioMysteryBench (human-difficult)49.7%estimated ± 6.8 pp, low confidence
DeepSearchQA91.6%estimated ± 6.8 pp, low confidence
CyberGym86.7%estimated ± 7.3 pp, medium confidence
JobBench55.3%estimated ± 10.7 pp, low confidence
SkillsBench62.2%estimated ± 4.1 pp, medium confidence
WideResearch82.0%estimated ± 3.5 pp, high confidence
AA Agentic Index47.6%estimated ± 8.2 pp, medium confidence
MMLU-Pro85.7%estimated ± 2.5 pp, low confidence
FailureSensorIQ67.9%estimated ± 2.1 pp, high confidence
GPQA94.3%estimated ± 1.6 pp, high confidence
Vals GPQA Diamond93.2%estimated ± 2.3 pp, high confidence
Vals MMLU-Pro88.4%estimated ± 1.8 pp, high confidence
Chess Puzzles59.0%measured
Gert Labs63.7%estimated ± 11.1 pp, low confidence
AA Harvey LAB v1.092.8%estimated ± 9.8 pp, low confidence
AA Tau3 Banking43.2%estimated ± 3.6 pp, medium confidence
ARC-AGI-194.5%measured
ARC-AGI-283.3%measured
EBR-bench48.0%estimated ± 12.7 pp, low confidence
FrontierMath v2 (Tier 4)58.5%measured
FrontierMath v2 (Tiers 1-3)82.5%measured
Furniture Assembly55.6%estimated ± 14.1 pp, low confidence
MaxxBench64.9%estimated ± 10.3 pp, low confidence
MazeBench7.2%estimated ± 6.4 pp, low confidence
Mystery Game Puzzles52.9%estimated ± 10.3 pp, low confidence
NYT Connections91.9%estimated ± 12.6 pp, low confidence
OSWorld-Verified83.9%estimated ± 5.7 pp, low confidence
CWE-Bench-AA54.9%estimated ± 6.4 pp, medium confidence
AA Cyber Index42.1%estimated ± 13.5 pp, low confidence
DeepsecBench-AA23.6%estimated ± 9.0 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)57.9%estimated ± 4.4 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)69.9%estimated ± 0.8 pp, medium confidence
AA ITBench49.8%estimated ± 4.1 pp, high confidence
DRACO85.2%estimated ± 2.9 pp, medium confidence
FrontierPhysics14.0%estimated ± 6.1 pp, medium confidence
HLE w/o tools42.7%measured
SpireClimb11.2%estimated ± 10.3 pp, low confidence
StealthMark59.0%estimated ± 9.7 pp, low confidence
Toolathlon-Verified74.1%estimated ± 7.4 pp, medium confidence
ARC-AGI-34.7%estimated ± 7.2 pp, medium confidence
SkillsBench official leaderboard (BenchFlow)59.2%estimated ± 6.2 pp, medium confidence
FACTS Multimodal50.5%estimated ± 5.6 pp, medium confidence
ExploitBench78.5%estimated ± 13.4 pp, low confidence
ExploitGym25.8%estimated ± 9.2 pp, low confidence
Claw-Eval80.0%estimated ± 6.4 pp, medium confidence
HMMT Feb 2026100.0%estimated ± 1.5 pp, low confidence
HMMT Nov 202595.9%estimated ± 1.5 pp, low confidence
MCP Atlas79.5%estimated ± 10.3 pp, low confidence
MMAnswerBench94.8%estimated ± 2.4 pp, low confidence
MMLU-Redux97.0%estimated ± 2.1 pp, medium confidence
SuperGPQA97.4%estimated ± 7.2 pp, low confidence
Toolathlon59.9%estimated ± 4.2 pp, high confidence
HealthBench Hard33.9%estimated ± 8.6 pp, low confidence
MedXpertQA (Text)60.2%estimated ± 8.9 pp, low confidence
ResearchClawBench19.9%estimated ± 2.3 pp, low confidence
DAYJOB9.4%estimated ± 6.3 pp, medium confidence
DAYJOB Healthcare6.8%estimated ± 6.1 pp, low confidence
BioMysteryBench (human-solvable)89.5%estimated ± 0.6 pp, medium confidence
HealthBench (raw)48.4%estimated ± 7.6 pp, low confidence
HealthBench (length-adjusted)56.9%estimated ± 2.5 pp, low confidence
HealthBench Professional52.0%estimated ± 4.2 pp, medium confidence
HLE-Verified52.2%estimated ± 1.3 pp, medium confidence
LABBench279.8%estimated ± 1.7 pp, low confidence
BlenderWorld (text-only)85.7%estimated ± 4.0 pp, medium confidence
Cookie-Bench88.6%estimated ± 2.5 pp, medium confidence
Agents' Last Exam39.5%estimated ± 11.6 pp, low confidence
SEC-Bench Pro69.3%estimated ± 12.3 pp, low confidence
BioBasicsBench34.3%estimated ± 5.1 pp, low confidence
LifeVal79.1%estimated ± 1.6 pp, medium confidence
ExploitBench Internal Port6.7%estimated ± 1.4 pp, medium confidence
PinchBench90.7%estimated ± 3.6 pp, low confidence