Published task benchmark

Return RMA Agent

Check policy and order state, create an eligible return, issue the label, and update the case.

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

Customer service and account operationsAnalysisEvaluated 9/11/2026100 evaluation examplesFallback pricing checked September 11, 2026
Analysis

Return RMA Agent

Check policy and order state, create an eligible return, issue the label, and update the case.

Customer service and account operationsPer return
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
631 in · 701 out · 10.96s p50
$8.73 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
31%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
568 in · 515 out · 6.68s p50
$65.70 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $0.2500 input · $1.00 output /1M
TrustedRouter · $0.3376 input · $0.9390 output /1M
43%
Claude Opus 4.7
Hosted · Off-the-shelf
844 in · 1,350 out · 16.46s p50
$3797 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $5.00 input · $25.00 output /1M
TrustedRouter · $5.28 input · $26.38 output /1M
73%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small13%100/100$26.95
Qwen3 235B A22Bmid38%100/100$175
DeepSeek V3mid43%100/100$65.70
Mistral Large 2407mid24%100/100$174
Arcee Trinity Large Thinkingmid25%100/100$117
GPT-5.4frontier67%100/100$1469
Claude Opus 4.7frontier73%100/100$3797
Gemini 3.1 Pro Previewfrontier67%100/100$1912
Gemma 4 E4B ITsmall31%100/100$8.73
Granite 4.1 8Bsmall15%100/100$8.96
Ministral 8B Instruct 2410small13%100/100$21.59
Qwen3 4B Instruct 2507small18%100/100$90.27

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-11T00:38:03.489Z.

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