benchgap
OpenAI · model

GPT-5.4 nano benchmark scores

As of 2026-11-22, GPT-5.4 nano (OpenAI) has measured scores on 36 benchmarks and estimated scores on 136 more.

BenchmarkScoreSource
AA-GPQA Diamond81.7%measured
AA-HLE28.3%measured
AA-SciCode47.2%measured
AIME2695.9%estimated ± 1.2 pp, medium confidence
Artificial Analysis Intelligence Index20.7%measured
CritPt9.3%measured
GDPval-AA22.4%measured
GPQA Diamond82.8%estimated ± 0.1 pp, medium confidence
HLE37.7%measured
IFBench74.2%estimated ± 6.6 pp, medium confidence
AA-LCR76.7%measured
LiveCodeBench v688.3%estimated ± 2.2 pp, medium confidence
AA-Omniscience Accuracy25.7%measured
SciCode44.7%estimated ± 2.9 pp, medium confidence
Terminal-Bench 2.138.7%estimated ± 6.7 pp, low confidence
BrowseComp79.9%estimated ± 4.2 pp, high confidence
HLE w/ tools56.6%estimated ± 4.1 pp, medium confidence
IFEval93.0%estimated ± 2.8 pp, high confidence
LongBench v263.0%estimated ± 5.5 pp, medium confidence
VITA-Bench37.7%estimated ± 9.9 pp, low confidence
τ²-bench results92.5%measured
AA Coding Index56.1%measured
AA-MMMU-Pro65.4%measured
APEX-Agents28.2%estimated ± 4.9 pp, low confidence
APEX-Agents-AA24.9%measured
BioMysteryBench (human-difficult)43.2%estimated ± 6.8 pp, low confidence
DeepSearchQA82.6%estimated ± 4.5 pp, high confidence
SWE-bench Verified76.6%estimated ± 4.4 pp, high confidence
BFCL v472.8%estimated ± 2.7 pp, high confidence
CyberGym59.2%estimated ± 5.3 pp, low confidence
JobBench27.2%estimated ± 7.3 pp, low confidence
SkillsBench44.6%estimated ± 9.1 pp, low confidence
SWE-bench Pro55.0%estimated ± 2.9 pp, medium confidence
WideResearch73.3%estimated ± 2.1 pp, low confidence
AA Agentic Index17.7%measured
MMLU-Pro84.8%estimated ± 2.5 pp, medium confidence
AA-IFBench75.9%measured
FailureSensorIQ66.8%estimated ± 2.1 pp, high confidence
GPQA82.8%measured
MMLU90.5%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond77.5%measured
Vals LiveCodeBench84.0%measured
Vals MMLU-Pro77.2%measured
MMMLU89.4%estimated ± 0.7 pp, medium confidence
MMMU76.6%estimated ± 0.8 pp, high confidence
Chess Puzzles45.0%measured
AA MMLU-Pro85.2%estimated ± 0.7 pp, medium confidence
Gert Labs41.7%estimated ± 9.3 pp, medium confidence
AA Harvey LAB v1.080.2%estimated ± 1.2 pp, low confidence
AA Tau3 Banking39.2%estimated ± 2.7 pp, medium confidence
AA Terminal-Bench 2.179.4%estimated ± 2.5 pp, medium confidence
ApprenticeBench2.7%estimated ± 11.0 pp, low confidence
ARC-AGI-151.5%measured
ARC-AGI-25.7%measured
CursorBench 3.137.1%estimated ± 3.4 pp, low confidence
CursorBench 3.241.0%estimated ± 4.4 pp, medium confidence
EBR-bench0.0%estimated ± 10.4 pp, low confidence
FrontierCode 1.1 Main11.1%estimated ± 4.3 pp, low confidence
FrontierMath v2 (Tier 4)12.2%measured
FrontierMath v2 (Tiers 1-3)44.9%measured
FrontierSWE v26.3%estimated ± 9.5 pp, low confidence
Furniture Assembly32.9%estimated ± 9.3 pp, low confidence
MaxxBench47.2%estimated ± 8.8 pp, low confidence
MazeBench0.4%estimated ± 5.5 pp, low confidence
Mystery Game Puzzles5.0%measured
NYT Connections85.4%estimated ± 12.6 pp, low confidence
OfficeQA Pro50.3%estimated ± 6.8 pp, low confidence
OSWorld-Verified39.0%measured
ParseBench-142.2%measured
Terminal-Bench 4.15.5%estimated ± 14.9 pp, low confidence
Terminal-Bench Hard43.8%estimated ± 3.9 pp, medium confidence
Vals SWE-bench69.8%measured
Terminal-Bench 2.1 (Vals AI)41.6%measured
VulcanBench v383.0%estimated ± 5.6 pp, low confidence
CWE-Bench-AA29.8%estimated ± 10.1 pp, low confidence
AA Cyber Index12.1%estimated ± 9.2 pp, low confidence
DeepsecBench-AA0.7%estimated ± 8.9 pp, low confidence
HealthBench Professional (length-adjusted, Artificial Analysis)52.1%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)53.8%estimated ± 0.8 pp, low confidence
AA ITBench38.5%estimated ± 4.0 pp, high confidence
Bug Hunt Bench5.8%estimated ± 9.0 pp, low confidence
CursorBench 4.030.8%estimated ± 3.4 pp, medium confidence
CWE-bench v131.6%estimated ± 6.4 pp, low confidence
DeepSWE24.7%estimated ± 6.4 pp, low confidence
DRACO70.3%estimated ± 2.0 pp, medium confidence
FrontierCode 2.026.1%estimated ± 9.3 pp, low confidence
FrontierPhysics2.5%estimated ± 5.0 pp, low confidence
HLE w/o tools24.3%measured
PostTrainBench v1.115.2%estimated ± 3.7 pp, medium confidence
SpireClimb0.0%estimated ± 9.3 pp, low confidence
StealthMark31.0%estimated ± 9.2 pp, low confidence
Toolathlon-Verified44.8%estimated ± 2.3 pp, low confidence
VulcanBench Frontier v477.4%estimated ± 6.8 pp, low confidence
GDP.pdf17.0%estimated ± 4.6 pp, medium confidence
AA Terminal-Bench 4.00.0%estimated ± 13.7 pp, low confidence
ARC-AGI-30.2%estimated ± 7.2 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
SkillsBench official leaderboard (BenchFlow)57.3%estimated ± 6.2 pp, medium confidence
CompileBench51.1%estimated ± 10.0 pp, low confidence
Vibe Code Bench26.1%measured
FACTS Multimodal45.4%estimated ± 5.6 pp, medium confidence
Hermes Bench53.2%estimated ± 2.9 pp, high confidence
Hermes Index20.8%estimated ± 8.7 pp, medium confidence
ParseBench-247.4%estimated ± 3.5 pp, high confidence
CharXiv77.2%estimated ± 5.6 pp, low confidence
CharXiv w/o tools73.9%estimated ± 1.2 pp, high confidence
ExploitBench10.6%estimated ± 7.2 pp, low confidence
ExploitGym0.5%estimated ± 10.6 pp, low confidence
AI2D_TEST88.0%estimated ± 2.0 pp, medium confidence
C-Eval94.8%estimated ± 0.9 pp, low confidence
Claw-Eval64.7%estimated ± 6.3 pp, low confidence
CountBench88.2%estimated ± 2.9 pp, low confidence
DeepPlanning37.9%estimated ± 13.1 pp, low confidence
DynaMath6.0%estimated ± 3.0 pp, low confidence
ERQA41.3%estimated ± 3.1 pp, medium confidence
HMMT Feb 202688.1%estimated ± 1.5 pp, medium confidence
HMMT Nov 202594.0%estimated ± 1.5 pp, medium confidence
MathVision74.1%estimated ± 3.0 pp, high confidence
MCP Atlas56.1%measured
MCP-Tasks76.0%estimated ± 0.6 pp, low confidence
MLVU (M-Avg)80.6%estimated ± 2.2 pp, high confidence
MMAnswerBench85.2%estimated ± 2.4 pp, medium confidence
MMLU-Redux94.2%estimated ± 0.8 pp, medium confidence
MMMU-Pro66.1%measured
NL2Repo40.1%estimated ± 4.7 pp, high confidence
PostTrainBench v119.0%estimated ± 3.1 pp, medium confidence
QwenClawBench54.9%estimated ± 2.6 pp, high confidence
RealWorldQA81.2%estimated ± 2.8 pp, high confidence
ScreenSpot Pro67.7%estimated ± 9.6 pp, low confidence
SimpleVQA52.7%estimated ± 7.5 pp, medium confidence
SuperGPQA77.7%estimated ± 7.2 pp, medium confidence
Terminal-Bench 2.046.3%measured
Toolathlon35.5%measured
V*0.5%estimated ± 7.5 pp, low confidence
Video-MME (w/o subtitle)77.7%estimated ± 1.3 pp, high confidence
Video-MME (with subtitle)81.0%estimated ± 3.0 pp, high confidence
VideoMMMU75.7%estimated ± 2.4 pp, high confidence
We-Math76.8%estimated ± 7.4 pp, low confidence
AA LiveCodeBench86.6%estimated ± 2.6 pp, high confidence
HealthBench Hard28.5%estimated ± 8.6 pp, low confidence
MedXpertQA (MM)41.2%estimated ± 2.4 pp, medium confidence
MedXpertQA (Text)39.3%estimated ± 8.9 pp, low confidence
React Native Evals73.0%estimated ± 3.0 pp, low confidence
ResearchClawBench16.9%estimated ± 1.8 pp, high confidence
SWE-Rebench52.2%estimated ± 6.5 pp, low confidence
PostTrainBench12.7%estimated ± 3.7 pp, medium confidence
DAYJOB1.3%estimated ± 3.2 pp, medium confidence
DAYJOB Healthcare0.7%estimated ± 4.2 pp, medium confidence
BioMysteryBench (human-solvable)82.8%estimated ± 0.6 pp, low confidence
HealthBench (raw)37.1%estimated ± 6.0 pp, low confidence
HealthBench (length-adjusted)49.1%estimated ± 3.4 pp, low confidence
HealthBench Professional33.9%estimated ± 4.2 pp, low confidence
HealthBench Professional (raw)48.4%estimated ± 4.5 pp, medium confidence
HLE-Verified0.0%estimated ± 6.8 pp, low confidence
LABBench274.8%estimated ± 2.4 pp, low confidence
ScarfBench25.2%estimated ± 9.1 pp, low confidence
BlenderWorld (text-only)70.0%estimated ± 2.8 pp, low confidence
Cookie-Bench76.9%estimated ± 1.3 pp, low confidence
STEMBench2.4%estimated ± 0.2 pp, low confidence
SimpleQA25.9%estimated ± 4.3 pp, medium confidence
BabyVision w/ Python88.8%estimated ± 3.8 pp, low confidence
OpenHarmony Bench v1.051.0%estimated ± 2.4 pp, high confidence
SEC-Bench Pro32.4%estimated ± 9.3 pp, low confidence
MMVU69.5%estimated ± 2.7 pp, medium confidence
PerceptionBench53.0%estimated ± 4.9 pp, low confidence
MRCRv285.1%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.3%estimated ± 2.6 pp, low confidence
OCRBench V267.6%estimated ± 3.3 pp, high confidence
BioBasicsBench33.9%estimated ± 2.8 pp, low confidence
MMMU-Pro w/ Python69.5%measured
ExploitBench Internal Port0.9%estimated ± 1.4 pp, medium confidence