Platform benchmark

claude-haiku-4-5

anthropic · MATH-500 · seed 42 · 2026-11-02 05:07 UTC

ResultsMATH-500 · n=96 · seed 42
85.4%

85.4% correct · 100.0% answered on MATH-500

Score per $
2.88
Run cost
$0.30
Latency p50
5.0 s
Latency p95
11.3 s
Items
96/96 completed
Dropped
0

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