breeze-7B-instruct: benchmark results
Provider: Other.
Aggregate Rating 1410 ± 50, rank #5595 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| TW-LegalBench - Statute Citation Accuracy | 2.6 | Statute accuracy (%): applicable provisions cited exactly (a | 68.2 |
| TW-LegalBench - Judgment Prediction | 38.3 | ROUGE-L x100 (0-100), Jieba-segmented, of the predicted judg | 63.6 |
| TW-LegalBench - Essay Questions | 10 | Rubric points judged fully correct (out of 580) across the 1 | 8.3 |
| TW-LegalBench - International Law | 46 | International-law questions: accuracy (%) on four-option sin | 8.3 |
| TW-LegalBench | 39.6 | Total (self-reported) | 0 |
| TW-LegalBench - Administrative Law | 37.4 | Administrative-law questions: accuracy (%) on four-option si | 0 |
| TW-LegalBench - Civil Law | 35.9 | Civil-law questions: accuracy (%) on four-option single-answ | 0 |
| TW-LegalBench - Constitutional Law | 46.2 | Constitutional-law questions: accuracy (%) on four-option si | 0 |
| TW-LegalBench - Criminal Law | 41.6 | Criminal-law questions: accuracy (%) on four-option single-a | 0 |
Interactive version: theaggregate.ai/model?slug=breeze-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.