Generation
Reply-Draft
Drafts a professional reply to an incoming email based on intent and tone.
OperationsPer email
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
243 in · 143 out · 2.19s p50
$2.02 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
97%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
229 in · 164 out · 2.25s p50
$23.13 /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
94%
GPT-5.4
Hosted · Off-the-shelf
228 in · 127 out · 1.79s p50
$248 /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 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 71% | 100/100 | $13.73 |
| Qwen3 235B A22B | mid | 87% | 100/100 | $41.50 |
| DeepSeek V3 | mid | 94% | 100/100 | $23.13 |
| Mistral Large 2407 | mid | 94% | 100/100 | $43.04 |
| Arcee Trinity Large Thinking | mid | 87% | 100/100 | $34.92 |
| GPT-5.4 | frontier | 99% | 100/100 | $248 |
| Claude Opus 4.7 | frontier | 96% | 100/100 | $931 |
| Gemini 3.1 Pro Preview | frontier | 96% | 100/100 | $1110 |
| Gemma 4 E4B IT | small | 97% | 100/100 | $2.02 |
| Granite 4.1 8B | small | 56% | 100/100 | $3.27 |
| Ministral 8B Instruct 2410 | small | 76% | 100/100 | $5.79 |
| Qwen3 4B Instruct 2507 | small | 71% | 100/100 | $14.58 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-06T15:48:02.631Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock