Public AI launches as Hugging Face Inference Provider

AI Impact Summary

Public AI has launched as a new Inference Provider on Hugging Face, offering serverless inference for public and sovereign AI models. This expands model accessibility through seamless integration with Hugging Face’s client SDKs and model pages, leveraging a vLLM-powered backend and a distributed infrastructure of partner-provided GPU resources. This allows developers to easily deploy and utilize models like Swiss AI Initiative’s Apertus-70B without managing infrastructure.