log inget an API key
Platform benchmark

gemma-4-31b-turbo-tee

google · GPQA-Diamond · seed 42 · 2026-10-01 06:46 UTC

ResultsGPQA-Diamond · n=96 · seed 42
75.0%

75.0% correct · 100.0% answered on GPQA-Diamond

Score per $
43.74
Run cost
$0.02
Latency p50
26.6 s
Latency p95
66.6 s
Items
96/96 completed
Dropped
0

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