benchgap
OpenAI · model

GPT-4 Turbo benchmark scores

As of 2026-11-22, GPT-4 Turbo (OpenAI) has measured scores on 3 benchmarks and estimated scores on 36 more.

BenchmarkScoreSource
AA-GPQA Diamond48.1%estimated ± 9.2 pp, medium confidence
AA-HLE3.1%measured
AA-SciCode32.0%estimated ± 3.4 pp, high confidence
Artificial Analysis Intelligence Index7.0%measured
CritPt0.1%estimated ± 3.4 pp, high confidence
GPQA Diamond67.9%estimated ± 2.4 pp, medium confidence
HLE16.4%estimated ± 8.5 pp, low confidence
LiveCodeBench v672.6%estimated ± 4.7 pp, high confidence
AA-Omniscience Accuracy17.7%estimated ± 7.3 pp, medium confidence
SciCode32.6%estimated ± 4.6 pp, high confidence
AA Coding Index21.5%measured
BioMysteryBench (human-difficult)30.1%estimated ± 8.3 pp, low confidence
SWE-bench Verified45.5%estimated ± 6.5 pp, medium confidence
SWE-bench Pro28.6%estimated ± 6.0 pp, low confidence
MMLU-Pro75.2%estimated ± 5.4 pp, medium confidence
GPQA63.9%estimated ± 7.8 pp, medium confidence
MMLU82.8%estimated ± 3.8 pp, high confidence
Vals GPQA Diamond49.8%estimated ± 5.7 pp, medium confidence
Vals LiveCodeBench55.8%estimated ± 7.1 pp, medium confidence
Vals MMLU-Pro71.7%estimated ± 4.2 pp, high confidence
MMMLU84.0%estimated ± 2.6 pp, low confidence
AA MMLU-Pro83.1%estimated ± 1.2 pp, medium confidence
ARC-AGI-171.9%estimated ± 7.9 pp, low confidence
ARC-AGI-226.1%estimated ± 14.4 pp, low confidence
Vals SWE-bench45.4%estimated ± 5.5 pp, medium confidence
HLE w/o tools21.5%estimated ± 4.9 pp, high confidence
ARC-AGI-31.3%estimated ± 9.8 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
Vibe Code Bench5.9%estimated ± 10.0 pp, low confidence
C-Eval87.7%estimated ± 1.0 pp, low confidence
MMLU-Redux87.6%estimated ± 1.2 pp, low confidence
NL2Repo12.8%estimated ± 4.7 pp, medium confidence
SuperGPQA61.7%estimated ± 10.1 pp, low confidence
AA LiveCodeBench84.0%estimated ± 4.7 pp, low confidence
React Native Evals69.5%estimated ± 3.2 pp, high confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional56.7%estimated ± 4.2 pp, low confidence
ScarfBench24.4%estimated ± 9.1 pp, low confidence
BioBasicsBench9.0%estimated ± 10.2 pp, low confidence