Summarization
Budget Variance Narrative
Given budget and actual line items with operational drivers, produce a concise variance explanation that correctly attributes amount, direction, and primary cause.
Finance & AccountingPer line item
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
353 in · 148 out · 2.26s p50
$2.31 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
93%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
320 in · 180 out · 1.97s p50
$27.70 /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
88%
GPT-5.4
Hosted · Off-the-shelf
313 in · 227 out · 2.57s p50
$419 /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
| Model | Tier | Quality | Judged | Scenario cost |
|---|---|---|---|---|
| Nemotron Nano 9B v2 | small | 54% | 100/100 | $24.48 |
| Qwen3 235B A22B | mid | 50% | 100/100 | $86.77 |
| DeepSeek V3 | mid | 88% | 100/100 | $27.70 |
| Mistral Large 2407 | mid | 60% | 100/100 | $78.49 |
| Arcee Trinity Large Thinking | mid | 69% | 100/100 | $82.36 |
| GPT-5.4 | frontier | 100% | 100/100 | $419 |
| Claude Opus 4.7 | frontier | 79% | 100/100 | $1889 |
| Gemini 3.1 Pro Preview | frontier | 98% | 100/100 | $1588 |
| Gemma 4 E4B IT | small | 93% | 100/100 | $2.31 |
| Granite 4.1 8B | small | 29% | 100/100 | $5.24 |
| Ministral 8B Instruct 2410 | small | 35% | 100/100 | $9.06 |
| Qwen3 4B Instruct 2507 | small | 69% | 100/100 | $22.73 |
LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-08T22:37:02.698Z.
Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.
drydock