Cohere now supported as Hugging Face Inference Provider on HF Hub

AI Impact Summary

Cohere is now available as an Inference Provider on Hugging Face Hub, enabling serverless inference for Cohere models within HF infrastructure. The lineup includes enterprise-focused options with 256k context, multilingual support, and retrieval-augmented generation with verifiable citations, plus tool-enabled workflows, expanding how these models can be deployed in document-heavy and multilingual environments. This creates a streamlined path for apps to route through the Cohere provider via HF client SDKs, reducing integration friction and potentially lowering latency for Cohere-based pipelines.