Platform benchmark

o4-mini

openai · GPQA-Diamond · seed 42 · 2026-07-17 11:20 UTC

ResultsGPQA-Diamond · n=96 · seed 42
77.1%

77.1% correct · 100.0% answered on GPQA-Diamond

Score per $
0.56
Run cost
$1.38
Latency p50
10.9 s
Latency p95
72.5 s
Items
96/96 completed
Dropped
0