Claude Fable 5

NewClaude 5

Ranks #5 of 35 on reading real contracts into structured billing data, +0.3 pts vs GPT-5.5.

76.3%

Accuracy

#5 of 35

2.3%

Hallucination

HIGH-confidence & wrong

$740

Cost / 1k contracts

$0.74 each · via OpenRouter

71.8s

Median latency

p90 96s

±0

Run-to-run σ

1 runs

0.1

Value

acc. pts per $/1k

Tokens per contract

Input36,488

Output7,494

Reasoning2,065

Reliability 100% · valid structured output across 30 calls.

Want the full picture? How we score