Platform benchmark

claude-fable-5-1

anthropic · GPQA-Diamond · seed 42 · 2026-10-01 05:06 UTC

ResultsGPQA-Diamond · n=96 · seed 42
90.6%

90.6% correct · 94.8% answered on GPQA-Diamond

Score per $
0.23
Run cost
$3.96
Latency p50
9.0 s
Latency p95
32.3 s
Items
96/96 completed
Dropped
0

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