Published task benchmark

Delivery Exception Agent

Investigate a delayed, lost, or damaged delivery and choose replacement, refund, or escalation.

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

Delivery Exception Agent

Investigate a delayed, lost, or damaged delivery and choose replacement, refund, or escalation.

Customer service and account operationsPer delivery
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
658 in · 708 out · 11.08s p50
$8.86 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
54%
Cheapest
Mistral Large 2407
123B · Off-the-shelf
707 in · 874 out · 24.90s p50
$176 /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
67%
GPT-5.4
Hosted · Off-the-shelf
557 in · 861 out · 9.17s p50
$1431 /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
93%
Best accuracy
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small27%100/100$34.58
Qwen3 235B A22Bmid44%100/100$224
DeepSeek V3mid56%100/100$80.25
Mistral Large 2407mid67%100/100$176
Arcee Trinity Large Thinkingmid46%100/100$122
GPT-5.4frontier93%100/100$1431
Claude Opus 4.7frontier86%100/100$4401
Gemini 3.1 Pro Previewfrontier63%100/100$2197
Gemma 4 E4B ITsmall54%100/100$8.86
Granite 4.1 8Bsmall29%100/100$10.49
Ministral 8B Instruct 2410small34%100/100$23.05
Qwen3 4B Instruct 2507small23%100/100$100

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-11T02:06:02.749Z.

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