aya-expanse-8B — benchmark results

Cohere Aya Expanse 8B multilingual model. Provider: Cohere. Released 2024-10-24. Access: Open.

Unified ELO 1455 ± 7, rank #972 of 1776 rated models, from 322 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
LatamBoard - FLORES Bidirectional46.97Score (%)97
LatamBoard - Translation Score48.24Score (%)97
LA Leaderboard - XNLI Galician49.77Accuracy (%)95.6
EuroEval Dutch NLU - DBRD91.72Sentiment classification Score (%)89.5
WMT25 English-Maasai ESA6ESA Human Score88.2
EuroEval French NLU - Allocine95.12Sentiment classification Score (%)88
MT-Bench PL - Roleplay9.25Judge Score (0-10)87.8
MT-Bench PL - STEM9.75Judge Score (0-10)87.8
EuroEval Spanish NLU - Sentiment Headlines ES49.16Sentiment classification Score (%)87.4
EuroEval English NLU - SST-568.72Sentiment classification Score (%)86.5
EuroEval Finnish NLU - Tydiqa FI71.14Reading comprehension Score (%)86.5
MT-Bench PL - Writing9.3Judge Score (0-10)85.7

Interactive version: theaggregate.ai/model?slug=aya-expanse-8b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.