Llama-3.1-Swallow-8B-v0.1: benchmark results
Provider: Institute of Science Tokyo. Released 2024-10-08. Access: Open.
Aggregate Rating 1506 ± 37, rank #2255 of 10916 rated models, from 54 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - ALT E TO J Bert Score JA F1 | 87.45 | Score (%) | 96.4 |
| Open Japanese LLM - ALT E TO J Bleu JA | 14.18 | Score (%) | 92.3 |
| Open Japanese LLM - Wiki Reading Char F1 | 87.36 | Score (%) | 90.6 |
| Open Japanese LLM - Wikicorpus J TO E Bert Score EN F1 | 91.05 | Score (%) | 90.3 |
| Open Japanese LLM - AIO Char F1 | 79.72 | Score (%) | 88.9 |
| Open Japanese LLM - Niilc Char F1 | 58.78 | Score (%) | 86.1 |
| Open Japanese LLM - Wikicorpus J TO E Bleu EN | 11.98 | Score (%) | 85.9 |
| pfgen-bench - Completion Mode - Fluency | 0.79 | Fluency Score | 85.9 |
| Open Japanese LLM - QA | 61.64 | Score (%) | 84.6 |
| Open Japanese LLM - Wikicorpus E TO J Bert Score JA F1 | 83.59 | Score (%) | 84.4 |
| pfgen-bench - Completion Mode - Score | 0.67 | pfgen Score (mean of three) | 84.3 |
| pfgen-bench - Completion Mode - Helpfulness | 0.35 | Helpfulness Score | 83.7 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-swallow-8b-v0-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.