Google Cloud, Hugging Face partner for AI development

393

Google Cloud has announced a significant collaboration forging a strategic partnership with the innovative startup Hugging Face to streamline and enhance the process of artificial intelligence (AI) software development within the company.

The companies reached a revenue-sharing agreement, but did not disclose the specific terms.

This partnership will allow developers utilizing Google Cloud’s technical infrastructure to seamlessly access and deploy Hugging Face’s repository of open-source AI software, including models and datasets.

Hugging Face has emerged as a prominent hub for open-source AI software, solidifying its position as a key player in the industry. Hugging Face entered into a similar partnership with Amazon.com’s cloud division in February last year.

The collaboration materialized due to the evolving preferences of cloud customers who now exhibit a greater inclination towards customizing or developing their own AI models instead of relying on pre-packaged solutions.

After the release of consumer-focused generative AI tools by Microsoft and Google’s Alphabet, businesses, both large and small, are keen on utilizing AI to enhance products or meet internal needs. This led to the partnership to develop task-specific AI software.

Also Read: Google to help fund Israeli AI firms, Palestinian startups

“It shows we are moving into a world where there’s not just one model, but there as many models as companies. Every company is training, fine tuning and optimizing their own models, which is the direction that the field is taking.” Hugging Face Chief Executive Clem Delangue said.

The surging demand for cloud-based AI computing implies that, if it maintains its current trajectory, it is poised to surpass the traditional cloud software market in the future.

“It always starts infrastructure up because you first got to put in the machines that reflect the demand – and grow from there,” Google Cloud CEO Thomas Kurian said.

Illustrating the growing interest in AI software development, Hugging Face, based in New York, anticipates its repository of AI-related software to nearly double in size over the next four months—a milestone achieved in a fraction of the time it took to reach its current state, as highlighted by Delangue.

Developers will have access to advanced versions of of Google’s proprietary AI chips, known as tensor processing units. Furthermore, Google aims to swiftly complete the technical integration needed to incorporate Nvidia’s highly sought-after advanced AI chips, the H100, in a matter of “weeks,” not “months,” as indicated by Kurian.

Source Reuters

Comments are closed.