Published task benchmark

Dev-Link

Translates a customer's "broken" complaint into a technical JIRA ticket.

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

Customer SupportGenerationEvaluated 9/5/2026100 evaluation examplesFallback pricing checked September 11, 2026
Generation

Dev-Link

Translates a customer's "broken" complaint into a technical JIRA ticket.

Customer SupportPer complaint
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
224 in · 285 out · 4.45s p50
$3.48 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
99%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
210 in · 323 out · 7.55s p50
$37.42 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.3376 input · $0.9390 output /1M
OpenRouter · $0.2900 input · $1.14 output /1M
85%
GPT-5.4
Hosted · Off-the-shelf
210 in · 313 out · 2.93s p50
$522 /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$19.32
Qwen3 235B A22Bmid34%100/100$92.73
DeepSeek V3mid85%100/100$37.42
Mistral Large 2407mid71%100/100$73.53
Arcee Trinity Large Thinkingmid67%100/100$55.19
GPT-5.4frontier100%100/100$522
Claude Opus 4.7frontier91%100/100$2176
Gemini 3.1 Pro Previewfrontier99%100/100$1393
Gemma 4 E4B ITsmall99%100/100$3.48
Granite 4.1 8Bsmall38%100/100$5.26
Ministral 8B Instruct 2410small31%100/100$8.36
Qwen3 4B Instruct 2507small24%100/100$28.03

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-05T21:01:03.171Z.

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