Published task benchmark

Clause-Extractor

Extracts and classifies specific clause types (termination, IP, confidentiality) from a contract.

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

Legal & ComplianceExtractionEvaluated 9/10/2026100 evaluation examplesFallback pricing checked September 11, 2026
Extraction

Clause-Extractor

Extracts and classifies specific clause types (termination, IP, confidentiality) from a contract.

Legal & CompliancePer contract
Not benchmarked
No result on this tier
Token profile pending
/100K
Run this tier to add it
Gemma 4 E4B IT
E4B · Off-the-shelf
453 in · 360 out · 5.37s p50
$4.75 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
23%
Cheapest
Mistral Large 2407
123B · Off-the-shelf
501 in · 471 out · 12.06s p50
$101 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.5275 input · $1.58 output /1M
OpenRouter · $2.00 input · $6.00 output /1M
44%
GPT-5.4
Hosted · Off-the-shelf
429 in · 456 out · 2.97s p50
$791 /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
32%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small20%100/100$23.95
Qwen3 235B A22Bmid31%100/100$77.99
DeepSeek V3mid32%100/100$43.91
Mistral Large 2407mid44%100/100$101
Arcee Trinity Large Thinkingmid23%100/100$94.39
GPT-5.4frontier32%100/100$791
Claude Opus 4.7frontier32%100/100$2793
Gemini 3.1 Pro Previewfrontier24%100/100$1472
Gemma 4 E4B ITsmall23%100/100$4.75
Granite 4.1 8Bsmall16%100/100$5.23
Ministral 8B Instruct 2410small22%100/100$14.71
Qwen3 4B Instruct 2507small20%100/100$25.49

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T09:02:02.604Z.

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