AI BriefingCohere Now Available for Direct Inference on Hugging Face
·2025.04.16 09:00
Key point
Cohere's major LLM and multimodal models are now available for serverless inference on the Hugging Face Hub.
1 / 2
Details
Cohere is now officially supported as a Hugging Face Inference Provider. This marks the first case of a model creator sharing and serving their models directly on the Hub.
Users can instantly run Cohere's models via Serverless Inference through the Hugging Face web UI or client SDKs.
Key Supported Models:
- Command R Series: Optimized for enterprise RAG, agentic tool use, and high security. In particular,
c4ai-command-a-03-2025supports a 256k context length. - Aya Expanse Series: Multilingual-specialized models supporting 20+ languages, including Korean.
- Aya Vision Series: Multimodal models that perform OCR, image captioning, visual reasoning, and more.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.