Platform benchmark

gpt-oss-20b

openai · MATH-500 · seed 42 · 2026-11-03 09:05 UTC

ResultsMATH-500 · n=96 · seed 42
91.7%

91.7% correct · 100.0% answered on MATH-500

Score per $
18.91
Run cost
$0.05
Latency p50
9.6 s
Latency p95
72.0 s
Items
96/96 completed
Dropped
0

Dataset: MATH-500 · per-item question and response text available in the results.