Generating Custom Images and Variations with Amazon Bedrock

The Amazon Bedrock Image Playground is tailored for developers, data scientists, and businesses to explore generative AI models for realistic image creation, making it particularly beneficial for applications like advertising and presentations. It features a user-friendly interface, a diverse selection of high-performing models, and customization options, allowing users to fine-tune models to meet specific needs.

In this lab, you will explore the Amazon Nova Canvas model and a few of its capabilities. First, you will interact with the model using the Amazon Bedrock Image Playground to generate and edit an image. Then, you will invoke the model to generate image variations using the Amazon Bedrock runtime in a Jupyter Notebook.

Learning objectives

Upon completion of this intermediate-level lab, you will be able to:

• Generate and edit images using the Amazon Bedrock Image Playground and the Nova Canvas model
• Employ the Amazon Bedrock runtime to generate variations of images using references

Intended audience

• Cloud Architects
• Software Engineers

Prerequisites

Familiarity with the following will be beneficial but is not required:

• Amazon Bedrock
• Python

The following content can be used to fulfill the prerequisites:

• Invoking Amazon Bedrock Models Using the Bedrock Runtime and AWS Lambda

Updates

July 22nd, 2025 - Updated the lab to use the Nova Canvas model

December 16th, 2024 - Resolved an issue preventing the lab from deploying