TL;DR:
- Google Cloud and Hugging Face have partnered to facilitate AI model development.
- Developers can now create, train, and deploy AI models without a Google Cloud subscription.
- The collaboration provides cost-effective access to Google’s tensor processing units (TPU) and GPU supercomputers.
- Hugging Face hosts open-source foundation models like Meta’s Llama 2 and Stability AI’s Stable Diffusion.
- The platform houses over 350,000 models for developers to utilize.
- Valued at $4.5 billion, Hugging Face has secured significant funding, including contributions from Google, Amazon, and Nvidia.
- Hugging Face users can expect access to Vertex AI and Kubernetes for model development in 2024.
- Google emphasizes its commitment to supporting the open-source AI ecosystem through this partnership.
Main AI News:
In the fast-evolving world of AI, Google Cloud’s recent alliance with Hugging Face is nothing short of a game-changer. This groundbreaking partnership is democratizing AI model development by enabling developers to build, train, and deploy models sans the requirement of a Google Cloud subscription, thereby democratizing access to cutting-edge technology.
The true marvel of this collaboration lies in the accessibility it offers to Google’s tensor processing units (TPU) and GPU supercomputers, featuring thousands of Nvidia’s highly sought-after H100s. Hugging Face, renowned for housing foundational open-source models like Meta’s Llama 2 and Stability AI’s Stable Diffusion, plays a pivotal role in this venture. Moreover, the platform boasts an extensive array of databases dedicated to model training.
With over 350,000 models residing on the Hugging Face platform, it’s akin to coders sharing their code on GitHub. Valued at a staggering $4.5 billion, Hugging Face’s journey to success has seen support pouring in, with major players like Google, Amazon, and Nvidia collectively contributing $235 million in the past year.
Google has exciting plans in store for Hugging Face users, as they can soon leverage the AI app-building prowess of Vertex AI and the versatile Kubernetes engine for model training and fine-tuning. These offerings are poised to become available in the first half of 2024.
In a statement, Google reaffirmed its commitment to nurturing the open-source AI ecosystem through its partnership with Hugging Face. While some of Google’s models can be found on the platform, prized creations like Gemini, the driving force behind the chatbot Bard, and the text-to-image model Imagen, remain exclusive as closed-source models.
Conclusion:
Google’s partnership with Hugging Face marks a significant development in the AI market. This collaboration democratizes access to powerful AI infrastructure and accelerates the growth of open-source AI models. As developers gain cost-effective access to cutting-edge technology, it fosters innovation and expands opportunities for AI application development, which is poised to drive further growth in the AI market.
