log inget an API key
Platform benchmark

gpt-6-astra

openai · GPQA-Diamond · seed 42 · 2026-10-17 15:03 UTC

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

94.8% correct · 100.0% answered on GPQA-Diamond

Score per $
0.47
Run cost
$2.01
Latency p50
7.7 s
Latency p95
24.6 s
Items
96/96 completed
Dropped
0

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