Cohere Releases North Small Translate: A 218B MoE Translation Model That Scores 83.6 on WMT26 Across 50 Languages

Cohere has released North Small Translate, an open-weight mixture-of-experts model designed for machine translation across 50 languages. The model utilizes 25 billion active parameters out of 218 billion total and achieved a score of 83.6 on the company's WMT26 evaluation benchmark. While the model weights are available for non-commercial use, businesses can access the technology through Cohere's commercial platform.