Building Generative AI applications with Amazon Bedrock

Amazon Bedrock is one of the easiest ways to enable generative AI-powered features in your applications. It is a managed service from AWS that provides access to Amazon and third-party foundation models through an API. Titan (Amazon), Llama-2 (Meta), Jurassic-2 (AI21 Labs), Claude (Anthropic), Command and Embed (Cohere), and Stable Diffusion (Stability.ai) are some of the models available in Amazon Bedrock.

Set up

1. Login to AWS and navigate to Amazon Bedrock console.

2. Go to Model Access from left menu and click on Manage Model Access.

3. Select the models you are interested in and request access.

Using Bedrock and running inference on a model

You can use the console playground to explore and experiment the models. To run inference using APIs, use the below code snippet (AWS python boto3 sample) from your local Integrated Development Environment (IDE).

Amazon Bedrock makes it easy and simple to build your generative AI-based applications. Check out Bedrock here and start building your generative AI features today!