Published task benchmark

Training Transcript Action Summary

Given an internal training transcript, extract key rules, required actions, deadlines, owners, and knowledge-check questions.

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

HR & Internal OpsSummarizationEvaluated 9/9/2026100 evaluation examplesFallback pricing checked September 11, 2026
Summarization

Training Transcript Action Summary

Given an internal training transcript, extract key rules, required actions, deadlines, owners, and knowledge-check questions.

HR & Internal OpsPer transcript
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
511 in · 427 out · 6.74s p50
$5.58 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
90%
Cheapest
Mistral Large 2407
123B · Off-the-shelf
563 in · 485 out · 12.88s p50
$106 /100K
Best current price · TrustedRouter
Compare 2 router rates · observed 9/11/2026
TrustedRouter · $0.5275 input · $1.58 output /1M
OpenRouter · $2.00 input · $6.00 output /1M
97%
GPT-5.4
Hosted · Off-the-shelf
478 in · 561 out · 4.62s p50
$961 /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

ModelTierQualityJudgedScenario cost
Nemotron Nano 9B v2small71%100/100$26.77
Qwen3 235B A22Bmid31%100/100$134
DeepSeek V3mid96%100/100$55.33
Mistral Large 2407mid97%100/100$106
Arcee Trinity Large Thinkingmid94%100/100$70.56
GPT-5.4frontier100%100/100$961
Claude Opus 4.7frontier98%100/100$2353
Gemini 3.1 Pro Previewfrontier99%100/100$1957
Gemma 4 E4B ITsmall90%100/100$5.58
Granite 4.1 8Bsmall79%100/100$7.67
Ministral 8B Instruct 2410small76%100/100$13.24
Qwen3 4B Instruct 2507small77%100/100$41.34

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-09T05:54:02.764Z.

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