benchgap
Google · model

Gemini 1.5 Pro benchmark scores

As of 2026-11-22, Gemini 1.5 Pro (Google) has measured scores on 8 benchmarks and estimated scores on 44 more.

BenchmarkScoreSource
AA-GPQA Diamond58.9%measured
AA-HLE4.6%measured
AA-SciCode32.9%estimated ± 3.4 pp, high confidence
Artificial Analysis Intelligence Index7.9%measured
CritPt0.2%estimated ± 3.4 pp, high confidence
GPQA Diamond79.8%estimated ± 1.9 pp, medium confidence
HLE16.8%estimated ± 8.5 pp, medium confidence
LiveCodeBench v674.3%estimated ± 4.7 pp, high confidence
AA-Omniscience Accuracy18.4%estimated ± 6.5 pp, medium confidence
SciCode33.6%estimated ± 4.6 pp, high confidence
AA Coding Index23.6%measured
AA-MMMU-Pro55.0%measured
BioMysteryBench (human-difficult)31.6%estimated ± 8.3 pp, low confidence
SWE-bench Verified48.5%estimated ± 6.5 pp, medium confidence
SWE-bench Pro30.8%estimated ± 6.0 pp, medium confidence
MMLU-Pro79.0%estimated ± 4.8 pp, high confidence
GPQA62.4%estimated ± 4.5 pp, high confidence
MMLU86.3%estimated ± 1.2 pp, high confidence
Vals GPQA Diamond58.3%measured
Vals LiveCodeBench41.7%measured
Vals MMLU-Pro75.3%measured
MMMLU81.2%estimated ± 1.7 pp, medium confidence
MMMU69.1%estimated ± 2.7 pp, high confidence
AA MMLU-Pro81.7%estimated ± 1.1 pp, medium confidence
ARC-AGI-172.4%estimated ± 7.9 pp, low confidence
ARC-AGI-226.1%estimated ± 14.4 pp, low confidence
FrontierSWE v20.0%estimated ± 14.3 pp, low confidence
ParseBench-143.9%estimated ± 6.2 pp, medium confidence
Vals SWE-bench47.6%estimated ± 5.5 pp, medium confidence
VulcanBench v376.2%estimated ± 5.6 pp, low confidence
HLE w/o tools24.4%estimated ± 4.9 pp, medium 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 Bench6.7%estimated ± 10.0 pp, low confidence
ParseBench-240.8%estimated ± 6.2 pp, medium confidence
CharXiv66.5%estimated ± 6.3 pp, low confidence
AI2D_TEST88.0%estimated ± 2.0 pp, medium confidence
C-Eval87.9%estimated ± 1.0 pp, low confidence
MMLU-Redux88.6%estimated ± 1.2 pp, low confidence
MMMU-Pro55.7%estimated ± 1.5 pp, high confidence
NL2Repo14.5%estimated ± 4.7 pp, medium confidence
ScreenSpot Pro60.6%estimated ± 10.8 pp, low confidence
SuperGPQA61.7%estimated ± 10.1 pp, low confidence
Video-MME (w/o subtitle)72.8%estimated ± 1.7 pp, medium confidence
AA LiveCodeBench84.0%estimated ± 4.7 pp, low confidence
React Native Evals70.0%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
SimpleQA17.9%estimated ± 7.8 pp, medium confidence
BioBasicsBench10.1%estimated ± 10.2 pp, low confidence
MMMU-Pro w/ Python60.2%estimated ± 2.0 pp, medium confidence