log inget an API key
Platform benchmark

claude-fable-5

anthropic · GPQA-Diamond · seed 42 · 2026-11-10 05:07 UTC

ResultsGPQA-Diamond · n=96 · seed 42
89.6%

89.6% correct · 93.8% answered on GPQA-Diamond

Score per $
0.12
Run cost
$7.63
Latency p50
10.0 s
Latency p95
42.3 s
Items
96/96 completed
Dropped
0

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