Grok 4.5

NewGrok 4

Ranks #5 of 43 on reading real contracts into structured billing data — +1.9 pts vs GPT-5.5.

71.2%

Accuracy

#5 of 43

15.4%

Hallucination

HIGH-confidence & wrong

$93.3

Cost / 1k contracts

$0.09 each · via OpenRouter

99.1s

Median latency

p90 167s

±2.6

Run-to-run σ

3 runs

0.76

Value

acc. pts per $/1k

Tokens per contract

Input22,786

Output7,951

Reasoning3,709

Reliability 100% · valid structured output across 18 calls.

Want the full picture? How we score →