aya-23-8B: benchmark results
Cohere Aya 23 8B multilingual instruction-tuned model. Provider: Cohere. Released 2024-05-23. Access: Open.
Aggregate Rating 1379 ± 10, rank #1299 of 1656 rated models, from 312 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - SUM | 12.13 | Score (%) | 92.1 |
| Open Japanese LLM - Xlsum JA Rouge2 | 12.11 | Score (%) | 91.8 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.72 | Score (%) | 90.1 |
| SAHARA - Paraphrase | 31.68 | spBLEU-1K | 87.2 |
| Open Japanese LLM - Xlsum JA RougeLsum | 26.03 | Score (%) | 85.8 |
| Open Japanese LLM - Xlsum JA Rouge1 | 30.82 | Score (%) | 84.7 |
| EuroEval French NLU - Allocine | 94.35 | Matthews correlation coefficient (%) | 77.4 |
| Open PL LLM - PPC (multiple choice, 5-shot) | 76.8 | Accuracy (%) | 74 |
| Open PL LLM - CBD (generative, 5-shot) | 32.21 | Macro F1 (%) | 73.3 |
| Open Portuguese LLM - ASSIN2 RTE | 91.9 | Macro F1 (%) | 71.8 |
| Open Persian LLM - ARC Easy | 80.46 | Accuracy (%) | 70 |
| EuroEval Portuguese NLU - HAREM | 46.27 | Micro F1 excluding MISC (%) | 67.5 |
Interactive version: theaggregate.ai/model?slug=aya-23-8b · How It Works · Data refreshed daily, snapshot 2026-10-20.