us.amazon.nova-pro-v1:0: benchmark results
Provider: Other.
Aggregate Rating 1501 ± 40, rank #2339 of 10916 rated models, from 96 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Nejumi 4 - BFCL - Irrelevance Detection | 100 | Accuracy (%) | 99.7 |
| Nejumi 4 - JHumanEval | 59.51 | Sandbox pass rate (%) | 98.6 |
| Nejumi 4 - Toxicity - Violation Categories | 58.73 | Criteria met (%) | 98.6 |
| Nejumi 4 - jaster (2-shot) - JSQuAD | 81.96 | Character F1 (x100) | 96.6 |
| Nejumi 4 - GLP - Information Retrieval | 81.95 | Score (%) | 94.9 |
| Nejumi 4 - jaster (0-shot) - JSQuAD | 81.95 | Character F1 (x100) | 93.1 |
| Nejumi 4 - jaster (2-shot) - JSICK | 84 | Exact match (%) | 89.7 |
| Nejumi 4 - JBBQ (2-shot) - Accuracy | 93 | Accuracy (%) | 82.1 |
| Nejumi 4 - ALT - Toxicity | 84.16 | Score (%) | 81 |
| Nejumi 4 - MT-Bench (Japanese) - Extraction | 96 | Judge rating (1-10, x10) | 79.7 |
| Nejumi 4 - jaster (0-shot) - JSICK | 82 | Exact match (%) | 73.8 |
| Nejumi 4 - jaster (2-shot) - ALT Japanese-to-English | 89.96 | COMET wmt22 (x100) | 69.3 |
Interactive version: theaggregate.ai/model?slug=us-amazon-nova-pro-v1-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.