log inget an API key
Platform benchmark

gpt-oss-20b

openai · GPQA-Diamond · seed 42 · 2026-11-03 09:12 UTC

ResultsGPQA-Diamond · n=96 · seed 42
63.5%

63.5% correct · 100.0% answered on GPQA-Diamond

Score per $
5.13
Run cost
$0.12
Latency p50
25.2 s
Latency p95
170.4 s
Items
96/96 completed
Dropped
0

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