Amazon (NASDAQ:AMZN.com Inc)'s cloud division, Amazon Web Services (AWS), has announced a partnership with artificial intelligence startup Hugging Face. The collaboration aims to simplify the deployment of thousands of AI models using Amazon's custom Inferentia2 computing chips.
Hugging Face, valued at $4.5 billion, has become a key resource for AI researchers and developers to share and modify chatbots and other AI software. The company is backed by a number of tech giants, including Amazon, Google parent Alphabet Inc. and Nvidia (NASDAQ:NVDA) Corp.
The collaboration addresses a common challenge faced by developers who, after refining open-source AI models like Meta Platforms Inc.’s Llama 3, are looking to use those models in software applications. With the integration of Hugging Face and AWS, developers can now easily run these models on AWS’s Inferentia2 chip.
Jeff Boudier, head of product and growth at Hugging Face, emphasized the importance of efficiency, saying, “One thing that’s really important to us is efficiency – making sure that as many people as possible can run the models and that they can run them in the most cost-effective way.”
AWS is interested in attracting more AI developers to its cloud services by offering a cost-effective way to run AI. While Nvidia is known for its dominance in training AI models, AWS argues that its Inferentia2 chip is particularly good at inference, the process of running trained models, which could be more cost-effective in the long run. Matt Wood, head of AI products at AWS, explains the advantage of their chip: “You train these models maybe once a month. But you could be running inference against them tens of thousands of times an hour. That’s where Inferentia2 really shines.”
The move is expected to boost AWS's position in the cloud computing market by providing developers with a more efficient and cost-effective means to deploy AI models at scale.
Source: Investing
