Published task benchmark

Policy-Check

Checks if a refund request meets the company's terms and conditions.

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

Customer SupportClassificationEvaluated 9/5/2026100 evaluation examplesFallback pricing checked September 11, 2026
Classification

Policy-Check

Checks if a refund request meets the company's terms and conditions.

Customer SupportPer request
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
176 in · 50 out · 2.01s p50
$5.81 /100K
AWS Bedrock temporary proxy price
View token rates
AWS Bedrock · $0.1545 input · $0.6180 output /1M
79%
DeepSeek V3
Hosted · Off-the-shelf
163 in · 24 out · 0.53s p50
$7.46 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $0.2900 input · $1.14 output /1M
TrustedRouter · $0.3376 input · $0.9390 output /1M
73%
GPT-5.4
Hosted · Off-the-shelf
159 in · 36 out · 1.15s p50
$93.75 /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
91%
Best accuracy
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small24%100/100$12.72
Qwen3 235B A22Bmid62%100/100$14.56
DeepSeek V3mid73%100/100$7.46
Mistral Large 2407mid58%100/100$14.30
Arcee Trinity Large Thinkingmid58%100/100$98.34
GPT-5.4frontier91%100/100$93.75
Claude Opus 4.7frontier81%100/100$253
Gemini 3.1 Pro Previewfrontier88%100/100$675
Gemma 4 E4B ITsmall63%100/100$0.5950
Granite 4.1 8Bsmall43%100/100$1.23
Ministral 8B Instruct 2410small24%100/100$3.00
Qwen3 4B Instruct 2507small79%100/100$5.81

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-05T20:49:03.087Z.

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