Acropolis

Gemini 2.5 Flash Preview (05-20)

GoogleGrade C. Evidence grade C of A to D: thin.ProprietaryReleased 20 May 2025
Compare with peers

Benchmark appearances

Best verified run per board, with confidence intervals and Pareto frontier status.

BenchmarkRankScoreCostStatusObserved
HLE

Epoch AI Benchmarking Hub · Default

#33 / 56
11%[8.6% to 13.4%]UnknownStandardNot stated
ARC-AGI-2

Epoch AI Benchmarking Hub · Default

#230 / 299
2.5%$0.32StandardNot stated
ARC-AGI-2

Epoch AI Benchmarking Hub · Default

#232 / 299
2.2%$0.03StandardNot stated
ARC-AGI-2

Epoch AI Benchmarking Hub · Default

#233 / 299
2.1%$0.20StandardNot stated
ARC-AGI-2

Epoch AI Benchmarking Hub · Default

#240 / 299
2%$0.32StandardNot stated
ARC-AGI-2

Epoch AI Benchmarking Hub · Default

#246 / 299
1.7%$0.06StandardNot stated