Published task benchmark

Chapter Glossary Extraction

Given a textbook chapter, extract domain terms, source-grounded definitions, and the section in which each term is introduced.

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

Education & EdTechExtractionEvaluated 9/9/2026100 evaluation examplesFallback pricing checked September 11, 2026
Extraction

Chapter Glossary Extraction

Given a textbook chapter, extract domain terms, source-grounded definitions, and the section in which each term is introduced.

Education & EdTechPer chapter
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
375 in · 561 out · 9.16s p50
$6.71 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
43%
Cheapest
DeepSeek V3
Hosted · Off-the-shelf
355 in · 489 out · 4.39s p50
$57.90 /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
34%
GPT-5.4
Hosted · Off-the-shelf
365 in · 550 out · 3.53s p50
$916 /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
43%
All 12 model results and methodology

Generated 100/100 examples

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small4%100/100$31.85
Qwen3 235B A22Bmid15%100/100$111
DeepSeek V3mid34%100/100$57.90
Mistral Large 2407mid19%100/100$118
Arcee Trinity Large Thinkingmid20%100/100$106
GPT-5.4frontier43%100/100$916
Claude Opus 4.7frontier30%100/100$2264
Gemini 3.1 Pro Previewfrontier31%100/100$2105
Gemma 4 E4B ITsmall43%100/100$6.71
Granite 4.1 8Bsmall9%100/100$7.11
Ministral 8B Instruct 2410small12%100/100$12.88
Qwen3 4B Instruct 2507small3%100/100$35.95

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T09:59:02.938Z.

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