Published task benchmark

BOL-Scanner

Extracts carrier info, weight, and destination from Bills of Lading.

Compare measured quality, workload cost, and latency across model tiers. Results are task-specific and directional—not a general model ranking.

Document IntelligenceExtractionEvaluated 9/6/2026100 evaluation examplesFallback pricing checked September 11, 2026
Extraction

BOL-Scanner

Extracts carrier info, weight, and destination from Bills of Lading.

Document IntelligencePer bill of lading
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
263 in · 186 out · 6.83s p50
$15.56 /100K
AWS Bedrock temporary proxy price
View token rates
AWS Bedrock · $0.1545 input · $0.6180 output /1M
74%
Qwen3 235B A22B
235B (22B active) · Off-the-shelf
263 in · 184 out · 2.00s p50
$45.45 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $0.4550 input · $1.82 output /1M
TrustedRouter · $0.6330 input · $2.11 output /1M
84%
GPT-5.4
Hosted · Off-the-shelf
232 in · 162 out · 1.70s p50
$301 /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
87%
Best accuracy
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small13%100/100$24.49
Qwen3 235B A22Bmid84%100/100$45.45
DeepSeek V3mid55%100/100$22.49
Mistral Large 2407mid63%100/100$47.95
Arcee Trinity Large Thinkingmid57%100/100$58.60
GPT-5.4frontier87%100/100$301
Claude Opus 4.7frontier50%100/100$955
Gemini 3.1 Pro Previewfrontier54%100/100$1125
Gemma 4 E4B ITsmall59%100/100$2.63
Granite 4.1 8Bsmall45%100/100$2.65
Ministral 8B Instruct 2410small57%100/100$7.35
Qwen3 4B Instruct 2507small74%100/100$15.56

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-06T11:23:02.556Z.

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?

Describe it in Task Explorer. If it is not measured yet, you can request a benchmark and follow future results.

Explore another task