Bedrock Claude

The Bedrock Claude component leverages Anthropic's Claude models within Amazon Bedrock to deliver advanced language model capabilities. It supports conversational AI, text generation, and summarization, focusing on ethical and responsible AI usage.

How to Connect Bedrock Claude to ZBrain Flow?

To integrate Bedrock Claude with ZBrain Flow, follow these detailed steps:

Step 1: Select Bedrock Claude as Your Connection

• Click on Bedrock Claude in ZBrain Flow.

Step 2: Configure the Bedrock Claude Connection

To establish a connection between Bedrock Claude and ZBrain Flow, enter the following details:

• Connection Name

  • Provide a unique and recognizable name for the connection.

• AWS Access Key

  • Sign in to the AWS Management Console and open the IAM (Identity and Access Management) console.

  • Navigate to the Users section in the left-hand menu.

  • Select the IAM user you want to use for Bedrock Claude integration.

  • Go to the Security Credentials tab.

  • Locate the Access Keys section and retrieve the Access Key ID.

  • If no access key exists, create a new one and make note of the corresponding Secret Access Key.

• AWS Secret Key

To find or generate a new AWS Secret Key:

• Sign in to the AWS Management Console and open the IAM console.

  1. Navigate to Users and select the IAM user you want to create a key for.

  2. Under the Security Credentials tab, locate the AWS Secret Key section.

  3. Click on Create AWS Secret Key.

  4. The Access Key ID and AWS Secret Key will be displayed.

  5. Store the AWS Secret Key securely, as it will not be shown again.

• Session Token

  • Temporary credentials can be used with Session Token.

• Region

  • Choose the appropriate geographical region where the AWS resources are hosted. By default, the region is set to us-east-1.

• Save the Connection

  • Click on Save to complete the setup and establish the connection.

Additional Settings

• Model Selection : Specify the Claude model variant to utilize within Bedrock, some models are suitable for natural language tasks, others specialize in code.

• System Prompt: Configure a predefined prompt that will guide the AI in generating consistent and relevant responses.

• Temperature: Adjust this setting to control the variability of the AI’s responses; lower values result in more consistent and predictable outputs.

• Maximum Tokens: Specify the maximum number of tokens the AI can use to construct its response for concise output.

• Question: Input the primary question or statement that will direct the AI’s response, ensuring it is specific to the task at hand.

• Image URLs: List of image URLs to be used as input for the model. Provide URLs separated by commas (e.g., url1, url2).

• Roles: This includes an array of roles to specify more accurate responses.

Trigger Available in ZBrain Flow

• New File Upload: Activates when a new file is uploaded or when a text-based input is received.