Platform benchmark

deepseek-v3.2-tee

deepseek · MATH-500 · seed 42 · 2026-11-10 05:54 UTC

ResultsMATH-500 · n=96 · seed 42
94.8%

94.8% correct · 100.0% answered on MATH-500

Score per $
8.26
Run cost
$0.11
Latency p50
17.8 s
Latency p95
97.7 s
Items
96/96 completed
Dropped
0

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