Llama-3.1-Swallow-70B-Instruct-v0.1: benchmark results
Provider: Institute of Science Tokyo. Released 2024-10-08. Access: Open.
Aggregate Rating 1579 ± 28, rank #1261 of 10916 rated models, from 58 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - ALT E TO J Bert Score JA F1 | 87.6 | Score (%) | 98.1 |
| Open Japanese LLM - ALT E TO J Comet Wmt22 | 91.57 | Score (%) | 98.1 |
| Open Japanese LLM - ALT J TO E Comet Wmt22 | 89.24 | Score (%) | 98 |
| Open Japanese LLM - ALT J TO E Bert Score EN F1 | 96.03 | Score (%) | 97.5 |
| pfgen-bench - Completion Mode - Truthfulness | 0.93 | Truthfulness Score | 97.4 |
| Open Japanese LLM - ALT J TO E Bleu EN | 19.3 | Score (%) | 96.9 |
| Open Japanese LLM - Wiki Reading Char F1 | 90.53 | Score (%) | 96.9 |
| Open Japanese LLM - AIO Char F1 | 86.06 | Score (%) | 96.4 |
| Open Japanese LLM - Wikicorpus J TO E Bleu EN | 12.92 | Score (%) | 96.4 |
| Open Japanese LLM - QA | 71.31 | Score (%) | 96.2 |
| Open Japanese LLM - MT | 85.65 | Score (%) | 95.9 |
| Open Japanese LLM - Commonsensemoralja Exact Match | 92.96 | Score (%) | 95.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-1-swallow-70b-instruct-v0-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.