Published task benchmark

B2B Catalog Schema Mapping

Given a supplier spreadsheet sample and a target product schema, map source columns, propose type conversions, and identify unmapped required fields.

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

E-commerce & LogisticsAnalysisEvaluated 9/9/2026100 evaluation examplesFallback pricing checked September 11, 2026
Analysis

B2B Catalog Schema Mapping

Given a supplier spreadsheet sample and a target product schema, map source columns, propose type conversions, and identify unmapped required fields.

E-commerce & LogisticsPer catalog
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
829 in · 1,690 out · 27.79s p50
$19.58 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
32%
DeepSeek V3
Hosted · Off-the-shelf
773 in · 1,470 out · 17.15s p50
$164 /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
56%
Claude Opus 4.7
Hosted · Off-the-shelf
1,225 in · 1,970 out · 20.51s p50
$5538 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $5.00 input · $25.00 output /1M
TrustedRouter · $5.28 input · $26.38 output /1M
53%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small2%100/100$50.90
Qwen3 235B A22Bmid33%100/100$357
DeepSeek V3mid56%100/100$164
Mistral Large 2407mid33%100/100$268
Arcee Trinity Large Thinkingmid11%100/100$176
GPT-5.4frontier29%100/100$3181
Claude Opus 4.7frontier53%100/100$5538
Gemini 3.1 Pro Previewfrontier0%100/100$2560
Gemma 4 E4B ITsmall32%100/100$19.58
Granite 4.1 8Bsmall16%100/100$14.74
Ministral 8B Instruct 2410small2%100/100$30.38
Qwen3 4B Instruct 2507small4%100/100$136

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T11:38:02.484Z.

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