Gemma-2-Llama-Swallow-2B-pt-v0.1: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1402 ± 40, rank #5964 of 10916 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Swallow - Pre-trained Japanese - NIILC | 54.9 | Character F1 (%) | 61.2 |
| Swallow - Pre-trained Japanese - WMT20 En-Ja | 26.1 | BLEU | 57.1 |
| Swallow - Pre-trained Japanese - JEMHopQA | 50.9 | Character F1 (%) | 54.1 |
| Swallow - Pre-trained Japanese - Average | 42.1 | Average Score (%) | 44.9 |
| Swallow - Pre-trained Japanese - JCommonsenseQA | 83 | Accuracy (%) | 44.9 |
| Swallow - Pre-trained Japanese - JMMLU | 46.1 | Accuracy (%) | 42.9 |
| Swallow - Pre-trained Japanese - WMT20 Ja-En | 19.5 | BLEU | 42.9 |
| Swallow - Pre-trained Japanese - JHumanEval | 25.1 | Pass@1 (%) | 40.8 |
| Swallow - Pre-trained English - HumanEval | 28.6 | Pass@1 (%) | 36.7 |
| Swallow - Pre-trained Japanese - JSQuAD | 86.3 | Character F1 (%) | 35.7 |
| Swallow - Pre-trained English - Average | 42.7 | Average Score (%) | 32.7 |
| Swallow - Pre-trained English - MMLU | 53.8 | Accuracy (%) | 32.7 |
Interactive version: theaggregate.ai/model?slug=gemma-2-llama-swallow-2b-pt-v0-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.