AICohereabout 7 hours ago
Cohere Launches Embed 5 Retrieval Models
9 min read
Cohere released Embed 5, a family of enterprise embedding models, in Pro and Fast tiers. Pro scores 85.8 on ViDoRe V3, beating Voyage 4 Large and Gemini Embedding 2, while Fast costs $0.08 per million tokens. Both share one embedding space, support 100+ languages and 128K context, and are available on Cohere's API, Microsoft Foundry, and Amazon SageMaker.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyCohere