Skip to main content

​
Atoti Intelligence Essentials

This is part of the Atoti Intelligence Essentials offer.
This guide explains how to configure Amazon Bedrock as the Large Language Model (LLM) provider for Atoti Intelligence in an Atoti Java project.

​
Prerequisites

Before configuring Amazon Bedrock, ensure the following requirements are met:
  • An AWS account with Amazon Bedrock access and appropriate permissions
  • The AWS region where Bedrock models are available

​
Add the dependency

Add the Spring AI Amazon Bedrock dependency to the project. Add the following to pom.xml:

​
Configure the model

Configure the Amazon Bedrock model in the application configuration file. This example assumes that AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY are passed as JVM arguments or environment variables. Add the following to application.yaml: 
spring:
  ai:
    bedrock:
      aws:
        access-key: ${AWS_ACCESS_KEY_ID}
        secret-key: ${AWS_SECRET_ACCESS_KEY}
        region: "your-aws-region"
      converse:
        chat:
          options:
            temperature: 0.
            maxTokens: 5000
            topP: 0.8
            model: eu.mistral.pixtral-large-2502-v1:0
Replace the following placeholders:
  • your-aws-region: the AWS region where Bedrock is available
  • Model name: choose an available Bedrock model
Refer to the Spring AI documentation for any additional configuration options.

​
Verify the configuration

After completing the configuration, verify that the LLM connection works:
  1. Start the Atoti application.
  2. Check the application logs for successful LLM initialization.
  3. Test an AI feature such as Visualize This or Auto-Explain.
After configuring Amazon Bedrock, proceed to set up AI features: