Analysis
Duplicate Engineering Ticket Detection
Given a new issue and an existing issue corpus, rank likely duplicates and distinguish same symptom, same root cause, and merely related requests.
Engineering & ProductPer ticket
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
818 in · 1,144 out · 17.37s p50
$15.53 /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
88%
Qwen3 235B A22B
235B (22B active) · Off-the-shelf
867 in · 1,492 out · 20.29s p50
$311 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $0.4550 input · $1.82 output /1M
TrustedRouter · $0.6330 input · $2.11 output /1M
90%
GPT-5.4
Hosted · Off-the-shelf
816 in · 1,345 out · 12.01s p50
$2222 /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
99%
Best accuracy
All 12 model results and methodology
Generated 92/100 examples — partial set accepted at the success threshold
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 54% | 92/92 | $48.37 |
| Qwen3 235B A22B | mid | 90% | 92/92 | $311 |
| DeepSeek V3 | mid | 89% | 92/92 | $125 |
| Mistral Large 2407 | mid | 76% | 92/92 | $231 |
| Arcee Trinity Large Thinking | mid | 60% | 92/92 | $149 |
| GPT-5.4 | frontier | 99% | 92/92 | $2222 |
| Claude Opus 4.7 | frontier | 95% | 92/92 | $5038 |
| Gemini 3.1 Pro Preview | frontier | 8% | 92/92 | $2567 |
| Gemma 4 E4B IT | small | 85% | 92/92 | $14.23 |
| Granite 4.1 8B | small | 88% | 92/92 | $15.53 |
| Ministral 8B Instruct 2410 | small | 34% | 92/92 | $28.47 |
| Qwen3 4B Instruct 2507 | small | 77% | 92/92 | $114 |
LLM-judge pass rate on 92 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T08:55:02.630Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock