log inget an API key
Platform benchmark

gemini-3.8-flash

gemini · GPQA-Diamond · seed 42 · 2026-10-17 06:22 UTC

ResultsGPQA-Diamond · n=96 · seed 42
94.8%

94.8% correct · 100.0% answered on GPQA-Diamond

Score per $
1.83
Run cost
$0.52
Latency p50
10.6 s
Latency p95
51.6 s
Items
96/96 completed
Dropped
0

Dataset: GPQA-Diamond · question and response text withheld under dataset access terms.