GPT-5.5

Ranks #6 of 43 on reading real contracts into structured billing data — +0 pts vs GPT-5.5.

69.3%

Accuracy

#6 of 43

14.8%

Hallucination

HIGH-confidence & wrong

$303

Cost / 1k contracts

$0.30 each · via OpenRouter

97.5s

Median latency

p90 124.2s

±2.1

Run-to-run σ

3 runs

0.23

Value

acc. pts per $/1k

Tokens per contract

Input21,576

Output6,505

Reasoning3,419

Reliability 100% · valid structured output across 18 calls.

Want the full picture? How we score →