Classification
Fixed-Asset Depreciation Classification
Given an asset description, placed-in-service date, jurisdiction, and supplied tax table, assign an asset class and calculate the depreciation schedule.
Finance & AccountingPer asset
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
421 in · 306 out · 3.31s p50
$5.17 /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
35%
Qwen3 235B A22B
235B (22B active) · Off-the-shelf
482 in · 295 out · 3.41s p50
$75.62 /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
75%
GPT-5.4
Hosted · Off-the-shelf
419 in · 458 out · 3.35s p50
$792 /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
88%
Best accuracy
All 12 model results and methodology
Generated 100/100 examples
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 8% | 100/100 | $34.04 |
| Qwen3 235B A22B | mid | 75% | 100/100 | $75.62 |
| DeepSeek V3 | mid | 74% | 100/100 | $30.17 |
| Mistral Large 2407 | mid | 51% | 100/100 | $84.61 |
| Arcee Trinity Large Thinking | mid | 41% | 100/100 | $135 |
| GPT-5.4 | frontier | 88% | 100/100 | $792 |
| Claude Opus 4.7 | frontier | 86% | 100/100 | $1801 |
| Gemini 3.1 Pro Preview | frontier | 80% | 100/100 | $1653 |
| Gemma 4 E4B IT | small | 10% | 100/100 | $3.14 |
| Granite 4.1 8B | small | 35% | 100/100 | $5.17 |
| Ministral 8B Instruct 2410 | small | 4% | 100/100 | $12.53 |
| Qwen3 4B Instruct 2507 | small | 22% | 100/100 | $45.64 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-08T22:36:02.980Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock