Prospect Trigger Summary
Given supplied company announcements and public posts, summarize recent buying signals, source each signal, and exclude unsupported personalization.
View token rates · observed 9/11/2026
Compare 2 router rates · observed 9/11/2026
Compare 2 router rates · observed 9/11/2026
All 12 model results and methodology
Generated 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 56% | 100/100 | $25.59 |
| Qwen3 235B A22B | mid | 80% | 100/100 | $126 |
| DeepSeek V3 | mid | 98% | 100/100 | $43.20 |
| Mistral Large 2407 | mid | 86% | 100/100 | $126 |
| Arcee Trinity Large Thinking | mid | 91% | 100/100 | $64.13 |
| GPT-5.4 | frontier | 100% | 100/100 | $867 |
| Claude Opus 4.7 | frontier | 96% | 100/100 | $2662 |
| Gemini 3.1 Pro Preview | frontier | 99% | 100/100 | $1798 |
| Gemma 4 E4B IT | small | 89% | 100/100 | $3.89 |
| Granite 4.1 8B | small | 71% | 100/100 | $7.18 |
| Ministral 8B Instruct 2410 | small | 84% | 100/100 | $13.80 |
| Qwen3 4B Instruct 2507 | small | 86% | 100/100 | $29.91 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T03:30:02.599Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock