Extraction

Given a dealer application, extract legal entity, principals, references, requested limit, tax identifiers, and missing fields into typed JSON.

E-commerce & LogisticsPer application

Not benchmarked

No result on this tier

Token profile pending

— /100K

Run this tier to add it

—

Qwen3 4B Instruct 2507

4B · Off-the-shelf

904 in · 634 out · 24.75s p50

$53.15 /100K

AWS Bedrock temporary proxy price

View token rates

AWS Bedrock · $0.1545 input · $0.6180 output /1M

DeepSeek V3

Hosted · Off-the-shelf

847 in · 572 out · 7.36s p50

$82.30 /100K

Best current price · TrustedRouter

Compare 2 router rates · observed 9/11/2026

TrustedRouter · $0.3376 input · $0.9390 output /1M

OpenRouter · $0.2900 input · $1.14 output /1M

18%

GPT-5.4

Hosted · Off-the-shelf

811 in · 589 out · 4.22s p50

$1086 /100K

Best current price · OpenRouter

Compare 2 router rates · observed 9/11/2026

OpenRouter · $2.50 input · $15.00 output /1M

TrustedRouter · $2.64 input · $15.82 output /1M

50%

All 12 model results and methodology

Generated 100/100 examples

Model	Tier	Quality	Judged	Scenario cost
Nemotron Nano 9B v2	small	1%	100/100	$49.22
Qwen3 235B A22B	mid	15%	100/100	$157
DeepSeek V3	mid	18%	100/100	$82.30
Mistral Large 2407	mid	8%	100/100	$179
Arcee Trinity Large Thinking	mid	1%	100/100	$177
GPT-5.4	frontier	50%	100/100	$1086
Claude Opus 4.7	frontier	39%	100/100	$2792
Gemini 3.1 Pro Preview	frontier	6%	100/100	$2559
Gemma 4 E4B IT	small	4%	100/100	$9.59
Granite 4.1 8B	small	6%	100/100	$9.61
Ministral 8B Instruct 2410	small	4%	100/100	$23.45
Qwen3 4B Instruct 2507	small	7%	100/100	$53.15

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T11:10:02.661Z.

Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.

drydock

Have a neighboring workload?