aya-23-8B — benchmark results
Cohere Aya 23 8B multilingual instruction-tuned model. Provider: Cohere. Released 2024-05-23. Access: Open.
Unified ELO 1371 ± 14, rank #1337 of 1776 rated models, from 155 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Japanese LLM - SUM | 12.13 | Score (%) | 92.1 |
| Open Japanese LLM - Xlsum JA Bert Score JA F1 | 70.72 | Score (%) | 90.1 |
| Open Japanese LLM - Xlsum JA Rouge2 | 12.11 | Score (%) | 90.1 |
| Open Japanese LLM - Xlsum JA RougeLsum | 26.03 | Score (%) | 85.3 |
| Open Japanese LLM - Xlsum JA Rouge1 | 30.82 | Score (%) | 84.5 |
| EuroEval French NLU - Allocine | 94.35 | Sentiment classification Score (%) | 77.4 |
| Open Persian LLM - ARC Easy | 80.46 | Accuracy (%) | 70 |
| EuroEval Portuguese NLU - HAREM | 46.27 | Named entity recognition Score (%) | 67.5 |
| Open CoT - LogiQA | 5.11 | CoT Gain (%) | 63 |
| EuroEval Dutch NLU - DBRD | 89.02 | Sentiment classification Score (%) | 61.9 |
| Open Persian LLM - ARC Challenge | 64.43 | Accuracy (%) | 60 |
| Open PL LLM - RAG | 60.76 | Average RAG Score (%) | 59.1 |
Interactive version: theaggregate.ai/model?slug=aya-23-8b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.