Cohere announced the release of Embed 5, a new family of embedding models for enterprise search and retrieval. The suite includes Embed 5 Pro, tuned for highest quality on multimodal, multilingual, financial, and code data, and Embed 5 Fast, built for lower latency and cost. Both models share a single embedding space, so indexes created with Pro can be queried with either model. The models are generally available on Cohere’s API, Microsoft Foundry, and Amazon SageMaker.
Why it matters
The models let companies improve search relevance while controlling inference costs.
