Skip links

Amazon Expands AI Capabilities of Bedrock with Three Models of Claude 3 by Anthropic

 Amazon has announced that its Bedrock service now includes all three versions of Anthropic’s Claude 3 artificial intelligence model—Opus, Sonnet, and Haiku. This makes Bedrock the first and only managed platform currently offering general access to all three Claude 3 models.

This development significantly enhances the capabilities of customers of all sizes to quickly test, create, and deploy generative AI applications across their organizations. As part of Amazon’s “three-layer” approach to AI, detailed in CEO Andy Jassy’s recent letter to shareholders, Bedrock provides foundational tools for developers and companies to build their own AI models.

On the intermediate level, Bedrock allows customers to use existing foundation models and tailor them for specific applications, offering a platform to scale generative AI applications through a variety of pre-trained models.

At the top level, Amazon is developing a range of generative AI applications across its consumer and AWS businesses, including AI-powered shopping assistants, enhanced voice assistants like Alexa, and new advertising solutions.

See also  Luma AI Startup Unveils Dream Machine Neural Network — It Generates Videos from Text Descriptions

Last month, Amazon announced an additional investment of $2.75 billion in Anthropic, bringing its total investment in the company to $4 billion. According to the agreement, Anthropic will use AWS as its primary cloud provider and will utilize AWS Tranium and Inferentia chips for its future AI models, and has committed to providing AWS customers with access to future generations of its foundation models on Bedrock.

“Generative AI is poised to be the most transformational technology of our time, and we believe our strategic collaboration with Anthropic will further enhance our customers’ experiences and look forward to what’s next,” said Dr. Swami Sivasubramanian, Vice President of Data and AI at Amazon Web Services (AWS).

Leave a comment

This website uses cookies to improve your web experience.
Explore
Drag