AIMarkTechPostabout 1 hour ago
Cohere Releases 218B MoE Translation Model
4 min read
Cohere released North Small Translate, an open-weight Mixture-of-Experts model with 218B total parameters and 25B active parameters. The model covers 50 languages and scores 83.6 on the WMT26 benchmark, outperforming DeepL, Google Translate, GLM 5.2, and Mistral Large 3.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyCohere
Summary by ByteBrief
