Published task benchmark

Policy-Puller

Identifies coverage limits, deductibles, and effective dates from insurance binders.

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

Policy-Puller

Identifies coverage limits, deductibles, and effective dates from insurance binders.

Document IntelligencePer binder
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
341 in · 328 out · 5.49s p50
$4.18 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
52%
Mistral Large 2407
123B · Off-the-shelf
378 in · 396 out · 10.43s p50
$82.61 /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
64%
Gemini 3.1 Pro Preview
Hosted · Off-the-shelf
326 in · 1,241 out · 8.43s p50
$1554 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $2.00 input · $12.00 output /1M
TrustedRouter · $2.11 input · $12.66 output /1M
74%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small15%100/100$32.28
Qwen3 235B A22Bmid63%100/100$69.43
DeepSeek V3mid53%100/100$37.73
Mistral Large 2407mid64%100/100$82.61
Arcee Trinity Large Thinkingmid41%100/100$70.69
GPT-5.4frontier64%100/100$527
Claude Opus 4.7frontier50%100/100$1353
Gemini 3.1 Pro Previewfrontier74%100/100$1554
Gemma 4 E4B ITsmall52%100/100$4.18
Granite 4.1 8Bsmall38%100/100$4.08
Ministral 8B Instruct 2410small38%100/100$10.08
Qwen3 4B Instruct 2507small35%100/100$22.46

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-06T11:14:02.554Z.

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