Published task benchmark

Court Transcript Fact Extractor

Pull dates, witnesses, and exhibits from trial transcripts.

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

Legal & ComplianceExtractionEvaluated 9/4/2026100 evaluation examplesFallback pricing checked September 11, 2026
Extraction

Court Transcript Fact Extractor

Pull dates, witnesses, and exhibits from trial transcripts.

Legal & CompliancePer 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
411 in · 299 out · 4.53s p50
$4.02 /100K
Current price · TrustedRouter
View token rates · observed 9/11/2026
TrustedRouter · $0.0211 input · $0.1055 output /1M
87%
Cheapest
Qwen3 235B A22B
235B (22B active) · Off-the-shelf
395 in · 282 out · 3.39s p50
$69.30 /100K
Best current price · OpenRouter
Compare 2 router rates · observed 9/11/2026
OpenRouter · $0.4550 input · $1.82 output /1M
TrustedRouter · $0.6330 input · $2.11 output /1M
98%
GPT-5.4
Hosted · Off-the-shelf
372 in · 295 out · 2.80s p50
$536 /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 v2small21%100/100$25.44
Qwen3 235B A22Bmid98%100/100$69.30
DeepSeek V3mid98%100/100$36.43
Mistral Large 2407mid97%100/100$73.90
Arcee Trinity Large Thinkingmid96%100/100$73.97
GPT-5.4frontier100%100/100$536
Claude Opus 4.7frontier98%100/100$1376
Gemini 3.1 Pro Previewfrontier99%100/100$1691
Gemma 4 E4B ITsmall87%100/100$4.02
Granite 4.1 8Bsmall79%100/100$4.34
Ministral 8B Instruct 2410small81%100/100$9.54
Qwen3 4B Instruct 2507small72%100/100$22.36

LLM-judge pass rate on 100 synthetic examples. Generator: gpt-5.2. Judge: gpt-5.2. Evaluated 2026-09-04T15:29:02.108Z.

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