How to beginner · 3 min read

How to use Claude on AWS Bedrock

Q: How to use Claude on AWS Bedrock

Use the boto3 client for bedrock-runtime to call Claude models on AWS Bedrock by specifying the model ID like anthropic.claude-3-5-sonnet-20241022-v2:0 and passing chat messages in the converse or invoke_model method. Parse the response JSON to extract the generated text from response["output"]["message"]["content"][0]["text"].

Quick answer

Use the boto3 client for bedrock-runtime to call Claude models on AWS Bedrock by specifying the model ID like anthropic.claude-3-5-sonnet-20241022-v2:0 and passing chat messages in the converse or invoke_model method. Parse the response JSON to extract the generated text from response["output"]["message"]["content"][0]["text"].

PREREQUISITES

Python 3.8+
AWS CLI configured with credentials
pip install boto3

Setup

Install the boto3 library and configure your AWS credentials with permissions to access Bedrock. Set your AWS region to one that supports Bedrock, such as us-east-1.

Run:

pip install boto3

Ensure your AWS credentials are configured via ~/.aws/credentials or environment variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY.

bash

pip install boto3

Step by step

This example shows how to send a chat completion request to the Claude model on AWS Bedrock using boto3. It uses the converse method with the model ID anthropic.claude-3-5-sonnet-20241022-v2:0 and extracts the assistant's reply.

python

import boto3

client = boto3.client('bedrock-runtime', region_name='us-east-1')

messages = [
    {"role": "user", "content": [{"type": "text", "text": "Hello Claude, how are you?"}]}
]

response = client.converse(
    modelId="anthropic.claude-3-5-sonnet-20241022-v2:0",
    messages=messages
)

# Extract the generated text
text = response['output']['message']['content'][0]['text']
print("Claude says:", text)

output

Claude says: Hello! I'm doing well, thank you. How can I assist you today?

Common variations

Using invoke_model method: You can also call invoke_model with a JSON body for more control.
Changing model: Replace modelId with other Claude versions available on Bedrock.
Streaming: Currently, Bedrock does not support streaming responses via boto3.

python

import json

body = json.dumps({
    "anthropic_version": "bedrock-2023-05-31",
    "max_tokens": 512,
    "messages": [
        {"role": "user", "content": [{"type": "text", "text": "Explain AI in simple terms."}]}
    ]
})

response = client.invoke_model(
    modelId="anthropic.claude-3-5-sonnet-20241022-v2:0",
    body=body
)

result = json.loads(response['body'])
text = result['output']['message']['content'][0]['text']
print("Claude explains:", text)

output

Claude explains: AI is a technology that allows machines to learn from data and perform tasks that usually require human intelligence.

Troubleshooting

AccessDeniedException: Ensure your AWS IAM user or role has Bedrock permissions.
Model not found: Verify the modelId is correct and available in your region.
Malformed request: Check the JSON structure of messages or body matches Bedrock API specs.

✅

Key Takeaways

Use boto3's bedrock-runtime client to call Claude models on AWS Bedrock.
Pass chat messages as JSON with role and content arrays per Bedrock API format.
Extract generated text from response['output']['message']['content'][0]['text'].
Ensure AWS credentials and permissions are properly configured for Bedrock access.
Use invoke_model for raw JSON requests and converse for structured chat calls.

Verified 2026-04 · anthropic.claude-3-5-sonnet-20241022-v2:0

Verify ↗