Set up the Amazon Bedrock API

This section describes how to set up your environment to make Amazon Bedrock API calls and provides examples of common use-cases. You can access the Amazon Bedrock API using the AWS Command Line Interface (AWS CLI), an AWS SDK, or a SageMaker Notebook.

Before you can access Amazon Bedrock APIs, you need to request access to the foundation models that you plan to use.

For details about the API operations and parameters, see the Amazon Bedrock API Reference.

The following resources provide additional information about the Amazon Bedrock API.

• AWS Command Line Interface

  • Amazon Bedrock CLI commands

  • Amazon Bedrock Runtime CLI commands

  • Agents for Amazon Bedrock CLI commands

  • Agents for Amazon Bedrock Runtime CLI commands

Add model access

Important

Before you can use any of the foundation models, you must request access to that model. If you try to use the model (with the API or within the console) before you have requested access to it, you will receive an error message. For more information, see Model access.

Amazon Bedrock endpoints

To connect programmatically to an AWS service, you use an endpoint. Refer to the Amazon Bedrock endpoints and quotas chapter in the AWS General Reference for information about the endpoints that you can use for Amazon Bedrock.

Amazon Bedrock provides the following service endpoints.

• bedrock – Contains control plane APIs for managing, training, and deploying models. For more information, see Amazon Bedrock Actions and Amazon Bedrock Data Types.

• bedrock-runtime – Contains data plane APIs for making inference requests for models hosted in Amazon Bedrock. For more information, see Amazon Bedrock Runtime Actions and Amazon Bedrock Runtime Data Types.

• bedrock-agent – Contains control plane APIs for creating and managing agents and knowledge bases. For more information, see Agents for Amazon Bedrock Actions and Agents for Amazon Bedrock Data Types.

• bedrock-agent-runtime – Contains data plane APIs for invoking agents and querying knowledge bases. For more information, see Agents for Amazon Bedrock Runtime Actions and Agents for Amazon Bedrock Runtime Data Types.

Setting up the AWS CLI

1. If you plan to use the CLI, install and configure the AWS CLI by following the steps at Install or update the latest version of the AWS Command Line Interface User Guide.

2. Configure your AWS credentials using the aws configure CLI command by following the steps at Configure the AWS CLI.

Refer to the following references for AWS CLI commands and operations:

• Amazon Bedrock CLI commands

• Amazon Bedrock Runtime CLI commands

• Agents for Amazon Bedrock CLI commands

• Agents for Amazon Bedrock Runtime CLI commands

Setting up an AWS SDK

AWS software development kits (SDKs) are available for many popular programming languages. Each SDK provides an API, code examples, and documentation that make it easier for developers to build applications in their preferred language. SDKs automatically perform useful tasks for you, such as:

• Cryptographically sign your service requests

• Retry requests

• Handle error responses

Refer to the following table to find general information about and code examples for each SDK, as well as the Amazon Bedrock API references for each SDK. You can also find code examples at Code examples for Amazon Bedrock using AWS SDKs.

SDK documentation	Code examples	Amazon Bedrock prefix	Amazon Bedrock runtime prefix	Agents for Amazon Bedrock prefix	Agents for Amazon Bedrock runtime prefix
AWS SDK for C++	AWS SDK for C++ code examples	bedrock	bedrock-runtime	bedrock-agent	bedrock-agent-runtime
AWS SDK for Go	AWS SDK for Go code examples	bedrock	bedrockruntime	bedrockagent	bedrockagentruntime
AWS SDK for Java	AWS SDK for Java code examples	bedrock	bedrockruntime	bedrockagent	bedrockagentruntime
AWS SDK for JavaScript	AWS SDK for JavaScript code examples	bedrock	bedrock-runtime	bedrock-agent	bedrock-agent-runtime
AWS SDK for Kotlin	AWS SDK for Kotlin code examples	bedrock	bedrockruntime	bedrockagent	bedrockagentruntime
AWS SDK for .NET	AWS SDK for .NET code examples	Bedrock	BedrockRuntime	BedrockAgent	BedrockAgentRuntime
AWS SDK for PHP	AWS SDK for PHP code examples	Bedrock	BedrockRuntime	BedrockAgent	BedrockAgentRuntime
AWS SDK for Python (Boto3)	AWS SDK for Python (Boto3) code examples	bedrock	bedrock-runtime	bedrock-agent	bedrock-agent-runtime
AWS SDK for Ruby	AWS SDK for Ruby code examples	Bedrock	BedrockRuntime	BedrockAgent	BedrockAgentRuntime
AWS SDK for Rust	AWS SDK for Rust code examples	aws-sdk-bedrock	aws-sdk-bedrockruntime	aws-sdk-bedrockagent	aws-sdk-bedrockagentruntime
AWS SDK for SAP ABAP	AWS SDK for SAP ABAP code examples	BDK	BDR	BDA	BDZ
AWS SDK for Swift	AWS SDK for Swift code examples	AWSBedrock	AWSBedrockRuntime	AWSBedrockAgent	AWSBedrockAgentRuntime

Using SageMaker notebooks

You can use the SDK for Python (Boto3) to invoke Amazon Bedrock API operations from a SageMaker notebook.

Configure the SageMaker role

Add Amazon Bedrock permissions to the IAM role that will use this SageMaker notebook.

From the IAM console, perform these steps:

1. Choose the IAM role, then choose Add Permissions and select Create Inline Policies from the dropdown list.

2. Include the following permission.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "bedrock:*",
            "Resource": "*"
        }
    ]
}           

Add the following permissions to the trust relationships.

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Principal": {
                "Service": "bedrock.amazonaws.com"
            },
            "Action": "sts:AssumeRole"
        },
        {
            "Sid": "",
            "Effect": "Allow",
            "Principal": {
                "Service": "sagemaker.amazonaws.com"
            },
            "Action": "sts:AssumeRole"
        }
    ]
}                   

Test the Runtime setup

Add the following code to your notebook and run the code.

import boto3
import json
bedrock = boto3.client(service_name='bedrock-runtime')

body = json.dumps({
    "prompt": "\n\nHuman:explain black holes to 8th graders\n\nAssistant:",
    "max_tokens_to_sample": 300,
    "temperature": 0.1,
    "top_p": 0.9,
})

modelId = 'anthropic.claude-v2'
accept = 'application/json'
contentType = 'application/json'

response = bedrock.invoke_model(body=body, modelId=modelId, accept=accept, contentType=contentType)

response_body = json.loads(response.get('body').read())
# text
print(response_body.get('completion'))

Test the Amazon Bedrock setup

Add the following code to your notebook and run the code.

import boto3
bedrock = boto3.client(service_name='bedrock')

bedrock.get_foundation_model(modelIdentifier='anthropic.claude-v2')