Google and Hugging Face collaborate to enhance AI services on Google Cloud Platform

Google and Hugging Face are teaming up to provide Hugging Face's AI services through Google Cloud Platform. The partnership aims to create an open platform for data scientists, machine learning engineers, and developers, using Hugging Face's open models and Google Cloud's advanced features. This collaboration will cover open science, open source, cloud, and hardware, allowing businesses to construct their own AI, while Hugging Face developers will also have access to Google Cloud's Vertex AI and secure infrastructure.

The collaboration aims to improve the efficiency and cost-effectiveness of training and deploying large AI models on Google Cloud, which will be achieved by facilitating the training, tuning, and serving of Hugging Face models with Vertex AI, and extending the reach of Cloud TPU v5e to more open source developers. The partnership is expected to support upcoming A3 VMs powered by NVIDIA H100 Tensor Core GPUs and integrate with Google Cloud Marketplace for simplified management and billing systems.

Deployment options for Vertex AI and GKE on the Hugging Face platform are not available just yet, but are expected to be in the first half of 2024. If you want to read the full statement, you can do so here.