Published task benchmark

Bill-Sieve

Extracts usage data (kWh, gallons) and billing cycles from utility statements.

Compare measured quality, workload cost, and latency across model tiers. Results are task-specific and directional—not a general model ranking.

Document IntelligenceExtractionEvaluated 9/6/2026100 evaluation examplesFallback pricing checked September 11, 2026
Extraction

Bill-Sieve

Extracts usage data (kWh, gallons) and billing cycles from utility statements.

Document IntelligencePer statement
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
366 in · 116 out · 1.51s p50
$2.00 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
17%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
309 in · 143 out · 1.30s p50
$23.86 /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
21%
GPT-5.4
Hosted · Off-the-shelf
307 in · 186 out · 1.80s p50
$356 /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
40%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small3%100/100$19.05
Qwen3 235B A22Bmid10%100/100$36.67
DeepSeek V3mid21%100/100$23.86
Mistral Large 2407mid18%100/100$60.61
Arcee Trinity Large Thinkingmid2%100/100$65.31
GPT-5.4frontier40%100/100$356
Claude Opus 4.7frontier37%100/100$926
Gemini 3.1 Pro Previewfrontier18%100/100$1151
Gemma 4 E4B ITsmall17%100/100$2.00
Granite 4.1 8Bsmall15%100/100$3.80
Ministral 8B Instruct 2410small2%100/100$9.84
Qwen3 4B Instruct 2507small2%100/100$17.64

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-06T11:01:02.735Z.

Directional: measured on a synthetic eval set generated by drydock. Cost/latency are not yet captured for taskrouter-run benchmarks.

drydock

Have a neighboring workload?

Describe it in Task Explorer. If it is not measured yet, you can request a benchmark and follow future results.

Explore another task