log inget an API key
Platform benchmark

kimi-k3

moonshot · GPQA-Diamond · seed 42 · 2026-10-01 07:31 UTC

ResultsGPQA-Diamond · n=96 · seed 42
95.8%

95.8% correct · 100.0% answered on GPQA-Diamond

Score per $
0.21
Run cost
$4.53
Latency p50
17.0 s
Latency p95
271.9 s
Items
96/96 completed
Dropped
0

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