3 min read
Cloud computing giant Amazon Web Services (AWS) said it is collaborating with startup Hugging Face, a software development hub, to make it easier to carry out artificial intelligence work (AI) in the cloud.
While new generative AI services like chat-based search engines from Microsoft and Google have captured the public’s imagination, tech companies such as AWS are also vying behind the scenes to supply the tools and services that software developers will need to weave similar technology into their products.
AWS, the biggest cloud computing provider, already offers tools to help developers create AI-based software, including proprietary computing chips for training AI algorithms on huge amounts of data at a lower cost than rivals to services that reduce how much time it takes to create a chatbot or other products.
This week AWS said it will work with Hugging Face, a New York-based company that has become a central place online where AI developers share open-source code and models. Clem Delangue, Hugging Face’s chief executive, said that while the deal is not exclusive, the startup is working closely with AWS on making it easy for developers the site and run it to take code from in the cloud.
“For this product collaboration, we’re dedicating significant engineering resources to build our shared products,” Delangue told Reuters in an interview. Delangue also said the next generation of Bloom, an open-source AI model that competes in size and scope with the model that Microsoft-backed OpenAI used to create ChatGPT, will be run on Trainium, a proprietary artificial intelligence chip created by AWS.
Swami Sivasubramanian, vice president of database, analytics, and machine learning at AWS, said he believes technologies like Trainium can help developers save money as AI takes up more computing power and AWS wants to make it less time-consuming for developers to adopt them.“We want to make sure they have access to our silicon and networking innovations,” Sivasubramanian told Reuters.
AWS already has more than 100,000 customers running AI applications in its cloud, Sivasubramanian said. These customers will now be able to access Hugging Face AI tools through Amazon’s SageMaker program. Additionally Hugging Face’s software developers can use Amazon’s cloud computing power and its chips designed for artificial intelligence tasks. The companies didn’t disclose the financial details of the partnership, but Amazon said it didn’t invest in the startup.
The partnership comes amid a flurry of agreements and investments that pair the largest cloud providers with companies working on generative AI. Last month, Microsoft Corp. struck a deal to invest in ChatGPT maker OpenAI that’s said to be valued at $10 billion, and is using the startup’s technology for Bing search. Earlier this month, Alphabet Inc.’s Google put nearly $400 million into OpenAI rival Anthropic, a person familiar with the matter said.
At stake is the ability to sell cloud-computing services to take advantage of a boom in interest in generative AI programs, which can create new text, photos, and graphics. Because the programs sort through a large volume of existing content to generate something new, they require considerable computing power delivered via the cloud and represent lucrative businesses to Amazon, Microsoft, and Google.
Amazon’s AWS division has a partnership with Stability AI already, which is best known for its stable Diffusion technology. It is a competitor to open AI’s Dall-E technology.
Read More:
