Published task benchmark

Delivery Exception Explanation

Given carrier scans, promised date, weather events, and exception codes, explain the current delay and next expected milestone without inventing an arrival time.

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

E-commerce & LogisticsSummarizationEvaluated 9/9/2026100 evaluation examplesFallback pricing checked September 11, 2026
Summarization

Delivery Exception Explanation

Given carrier scans, promised date, weather events, and exception codes, explain the current delay and next expected milestone without inventing an arrival time.

E-commerce & LogisticsPer delivery exception
Not benchmarked
No result on this tier
Token profile pending
/100K
Run this tier to add it
Granite 4.1 8B
8B · Off-the-shelf
387 in · 344 out · 4.61s p50
$5.38 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/4/2026
OpenRouter · $0.0500 input · $0.1000 output /1M
TrustedRouter · $0.0527 input · $0.1055 output /1M
74%
Mistral Large 2407
123B · Off-the-shelf
472 in · 303 out · 7.94s p50
$72.85 /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
100%
Best accuracy
GPT-5.4
Hosted · Off-the-shelf
381 in · 226 out · 2.83s p50
$434 /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
100%
Best accuracy
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small49%100/100$17.64
Qwen3 235B A22Bmid69%100/100$70.80
DeepSeek V3mid85%100/100$25.88
Mistral Large 2407mid100%100/100$72.85
Arcee Trinity Large Thinkingmid67%100/100$91.95
GPT-5.4frontier100%100/100$434
Claude Opus 4.7frontier90%100/100$1971
Gemini 3.1 Pro Previewfrontier95%100/100$1505
Gemma 4 E4B ITsmall69%100/100$2.08
Granite 4.1 8Bsmall74%100/100$5.38
Ministral 8B Instruct 2410small63%100/100$9.84
Qwen3 4B Instruct 2507small47%100/100$20.27

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T10:08:03.100Z.

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