log inget an API key
Platform benchmark

claude-fable-5

anthropic · GPQA-Diamond · seed 42 · 2026-11-02 05:04 UTC

ResultsGPQA-Diamond · n=96 · seed 42
92.7%

92.7% correct · 95.8% answered on GPQA-Diamond

Score per $
0.25
Run cost
$3.74
Latency p50
9.9 s
Latency p95
34.6 s
Items
96/96 completed
Dropped
0

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