llama-3-youko-8B-instruct: benchmark results
Rinna Youko 8B instruction release using supervised fine-tuning, a Llama 3 chat-vector addition that skips embeddings, and DPO. It adopts the Llama 3 chat format. Provider: rinna. Released 2024-07-25. Access: Open.
Aggregate Rating 1500 ± 28, rank #2357 of 10916 rated models, from 58 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - Wikicorpus J TO E Comet Wmt22 | 77.3 | Score (%) | 97.6 |
| Open Japanese LLM - Wikicorpus J TO E Bert Score EN F1 | 91.47 | Score (%) | 94.1 |
| Open Japanese LLM - Wikicorpus E TO J Comet Wmt22 | 84.86 | Score (%) | 91.2 |
| Open Japanese LLM - MT | 85.17 | Score (%) | 90.6 |
| Open Japanese LLM - Wikicorpus E TO J Bert Score JA F1 | 84.7 | Score (%) | 90.6 |
| Open Japanese LLM - Wikicorpus J TO E Bleu EN | 12.32 | Score (%) | 90.3 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.51 | Score (%) | 86.7 |
| Open Japanese LLM - SUM | 11.3 | Score (%) | 85.6 |
| Open Japanese LLM - Xlsum JA Rouge2 | 11.3 | Score (%) | 85.5 |
| Open Japanese LLM - Wikicorpus E TO J Bleu JA | 13.21 | Score (%) | 83.9 |
| Open Japanese LLM - Xlsum JA Bleu JA | 3.09 | Score (%) | 83.8 |
| Open Japanese LLM - Wiki Reading Char F1 | 83.74 | Score (%) | 83.2 |
Interactive version: theaggregate.ai/model?slug=llama-3-youko-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-08.