plamo-3-nict-8B-base: benchmark results
Provider: Other.
Aggregate Rating 1461 ± 39, rank #3252 of 10916 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Swallow - Pre-trained Japanese - JSQuAD | 92.6 | Character F1 (%) | 93.9 |
| Swallow - Pre-trained English - SQuAD2 | 64.5 | Exact Match (%) | 89.8 |
| Swallow - Pre-trained Japanese - WMT20 En-Ja | 30.1 | BLEU | 88.8 |
| Swallow - Pre-trained Japanese - NIILC | 64.5 | Character F1 (%) | 81.6 |
| Swallow - Pre-trained Japanese - JHumanEval | 38.4 | Pass@1 (%) | 61.2 |
| Swallow - Pre-trained Japanese - JEMHopQA | 52.4 | Character F1 (%) | 59.2 |
| Swallow - Pre-trained Japanese - JMMLU | 57.3 | Accuracy (%) | 59.2 |
| Swallow - Pre-trained Japanese - WMT20 Ja-En | 21 | BLEU | 59.2 |
| Swallow - Pre-trained Japanese - Average | 49.5 | Average Score (%) | 57.1 |
| Swallow - Pre-trained English - BBH | 56.8 | Exact Match (%) | 55.1 |
| Swallow - Pre-trained English - Average | 52.7 | Average Score (%) | 51 |
| Swallow - Pre-trained English - TriviaQA | 55.3 | Exact Match (%) | 51 |
Interactive version: theaggregate.ai/model?slug=plamo-3-nict-8b-base · How It Works · Data refreshed daily, snapshot 2026-11-08.