AWS customers can use Bedrock to write, build chatbots, summarize text, classify images and more based on text prompts. It gives its users a choice of Amazon’s Titan foundation model (FM) and several startups’ models, including Anthropic’s Claude (a Google-backed ChatGPT rival from former OpenAI employees), AI21’s Jurassic-2 (a language model specializing in Spanish, French, German, Portuguese, Italian and Dutch) and Stable Diffusion (a popular open-source image generator). Additionally, businesses and developers can customize how the models work based on input — which Amazon says won’t be used for training the models, according to CNBC. That should (theoretically) address a crucial privacy concern for businesses entering sensitive data.
Amazon views the range of AI models on offer as a way of providing flexibility to customers. The company’s description reads, “With Bedrock’s serverless experience, you can get started quickly, privately customize FMs with your own data, and easily integrate and deploy them into your applications using the AWS tools and capabilities you are familiar with (including integrations with Amazon SageMaker ML features like Experiments to test different models and Pipelines to manage your FMs at scale) without having to manage any infrastructure.”
“Most companies want to use these large language models, but the really good ones take billions of dollars to train and many years and most companies don’t want to go through that,” Amazon CEO Andy Jassy told CNBC on Thursday. “So what they want to do is they want to work off of a foundational model that’s big and great already and then have the ability to customize it for their own purposes. And that’s what Bedrock is.”
Amazon says C3.ai, Pegasystems, Accenture and Deloitte are some early businesses lined up to try Bedrock. The company hasn’t yet announced pricing for the AWS toolset, and it’s currently opening access through a waitlist. You can read more and apply for admission at the project’s website.
0 Commentaires