Gemini 3.5 Flash Lite

NewGemini 3

Ranks #6 of 35 on reading real contracts into structured billing data, +0.1 pts vs GPT-5.5.

76.1%

Accuracy

#6 of 35

21.8%

Hallucination

HIGH-confidence & wrong

$17.2

Cost / 1k contracts

$0.02 each · via OpenRouter

9.1s

Median latency

p90 12.5s

±0

Run-to-run σ

1 runs

4.43

Value

acc. pts per $/1k

Tokens per contract

Input26,048

Output3,750

Reasoning0

Reliability 100% · valid structured output across 30 calls.

Want the full picture? How we score