Making thousands of open LLMs bloom in the Vertex AI Model Garden

Hugging Face Blog · 2 years ago

Google and Hugging Face launched Deploy on Google Cloud, a tool that enables developers to deploy thousands of open-source language models to Google Cloud's Vertex AI or Kubernetes Engine directly from the Hugging Face Hub. Models tagged with "text-generation-inference" are immediately supported, with deployment achievable through a one-click process from either the Hugging Face model card or Google's Vertex Model Garden console. Developers can now build production generative AI applications without managing underlying infrastructure, with pre-configured hardware settings and secure deployment within their own Google Cloud accounts.