Published task benchmark

EBITDA Add-Back Identification

Given a synthetic general ledger and normalization policy, identify candidate nonrecurring expenses, calculate proposed add-backs, and provide the policy basis for each.

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

Finance & AccountingAnalysisEvaluated 9/10/2026100 evaluation examplesFallback pricing checked September 11, 2026
Analysis

EBITDA Add-Back Identification

Given a synthetic general ledger and normalization policy, identify candidate nonrecurring expenses, calculate proposed add-backs, and provide the policy basis for each.

Finance & AccountingPer general ledger
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
1,029 in · 1,521 out · 24.37s p50
$18.22 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
42%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
944 in · 1,613 out · 13.68s p50
$183 /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
69%
GPT-5.4
Hosted · Off-the-shelf
915 in · 1,853 out · 14.43s p50
$3008 /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
70%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small26%100/100$50.25
Qwen3 235B A22Bmid52%100/100$382
DeepSeek V3mid69%100/100$183
Mistral Large 2407mid50%100/100$322
Arcee Trinity Large Thinkingmid36%100/100$171
GPT-5.4frontier70%100/100$3008
Claude Opus 4.7frontier55%100/100$5684
Gemini 3.1 Pro Previewfrontier0%99/100$2598
Gemma 4 E4B ITsmall42%100/100$18.22
Granite 4.1 8Bsmall33%100/100$18.43
Ministral 8B Instruct 2410small17%100/100$34.47
Qwen3 4B Instruct 2507small26%100/100$133

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-10T09:00:02.785Z.

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