AWS Bedrock

Amazon Bedrock provides access to foundation models (FMs) from AI companies through a unified API. Build generative AI applications with text generation, embeddings, and image generation capabilities.

Table of Contents

• Core Concepts
• Common Patterns
• CLI Reference
• Best Practices
• Troubleshooting
• References

Core Concepts

Foundation Models

Pre-trained models available through Bedrock:

• Claude (Anthropic): Text generation, analysis, coding
• Titan (Amazon): Text, embeddings, image generation
• Llama (Meta): Open-weight text generation
• Mistral: Efficient text generation
• Stable Diffusion (Stability AI): Image generation

Model Access

Models must be enabled in your account before use:

• Request access in Bedrock console
• Some models require acceptance of EULAs
• Access is region-specific

Inference Types

Common Patterns

Invoke Claude (Text Generation)

AWS CLI:

aws bedrock-runtime invoke-model \
  --model-id anthropic.claude-3-sonnet-20240229-v1:0 \
  --content-type application/json \
  --accept application/json \
  --body '{
    "anthropic_version": "bedrock-2023-05-31",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Explain AWS Lambda in 3 sentences."}
    ]
  }' \
  response.json

cat response.json | python3 -c "import sys,json; print(json.load(sys.stdin)['content'][0]['text'])"

boto3:

import boto3
import json

bedrock = boto3.client('bedrock-runtime')

def invoke_claude(prompt, max_tokens=1024):
    response = bedrock.invoke_model(
        modelId='anthropic.claude-3-sonnet-20240229-v1:0',
        contentType='application/json',
        accept='application/json',
        body=json.dumps({
            'anthropic_version': 'bedrock-2023-05-31',
            'max_tokens': max_tokens,
            'messages': [
                {'role': 'user', 'content': prompt}
            ]
        })
    )

    result = json.loads(response['body'].read())
    return result['content'][0]['text']

# Usage
response = invoke_claude('What is Amazon S3?')
print(response)

Streaming Response

import boto3
import json

bedrock = boto3.client('bedrock-runtime')

def stream_claude(prompt):
    response = bedrock.invoke_model_with_response_stream(
        modelId='anthropic.claude-3-sonnet-20240229-v1:0',
        contentType='application/json',
        accept='application/json',
        body=json.dumps({
            'anthropic_version': 'bedrock-2023-05-31',
            'max_tokens': 1024,
            'messages': [
                {'role': 'user', 'content': prompt}
            ]
        })
    )

    for event in response['body']:
        chunk = json.loads(event['chunk']['bytes'])
        if chunk['type'] == 'content_block_delta':
            yield chunk['delta'].get('text', '')

# Usage
for text in stream_claude('Write a haiku about cloud computing.'):
    print(text, end='', flush=True)

Generate Embeddings

import boto3
import json

bedrock = boto3.client('bedrock-runtime')

def get_embedding(text):
    response = bedrock.invoke_model(
        modelId='amazon.titan-embed-text-v2:0',
        contentType='application/json',
        accept='application/json',
        body=json.dumps({
            'inputText': text,
            'dimensions': 1024,
            'normalize': True
        })
    )

    result = json.loads(response['body'].read())
    return result['embedding']

# Usage
embedding = get_embedding('AWS Lambda is a serverless compute service.')
print(f'Embedding dimension: {len(embedding)}')

Multi-Turn Conversation

import boto3
import json

bedrock = boto3.client('bedrock-runtime')

class Conversation:
    def __init__(self, system_prompt=None):
        self.messages = []
        self.system = system_prompt

    def chat(self, user_message):
        self.messages.append({
            'role': 'user',
            'content': user_message
        })

        body = {
            'anthropic_version': 'bedrock-2023-05-31',
            'max_tokens': 1024,
            'messages': self.messages
        }

        if self.system:
            body['system'] = self.system

        response = bedrock.invoke_model(
            modelId='anthropic.claude-3-sonnet-20240229-v1:0',
            contentType='application/json',
            accept='application/json',
            body=json.dumps(body)
        )

        result = json.loads(response['body'].read())
        assistant_message = result['content'][0]['text']

        self.messages.append({
            'role': 'assistant',
            'content': assistant_message
        })

        return assistant_message

# Usage
conv = Conversation(system_prompt='You are an AWS solutions architect.')
print(conv.chat('What database should I use for a chat application?'))
print(conv.chat('What about for time-series data?'))

List and Check Available Models

# List all foundation models
aws bedrock list-foundation-models \
  --query 'modelSummaries[*].[modelId,modelName,providerName]' \
  --output table

# Filter by provider
aws bedrock list-foundation-models \
  --by-provider anthropic \
  --query 'modelSummaries[*].modelId'

CLI Reference

Bedrock (Control Plane)

Bedrock Runtime (Data Plane)

Bedrock Agent Runtime

Best Practices

Cost Optimization

• Use appropriate models: Smaller models for simple tasks
• Set max_tokens: Limit output length when possible
• Cache responses: For repeated identical queries
• Batch when possible: Use batch inference for bulk processing
• Monitor usage: Set up CloudWatch alarms for cost

Performance

• Use streaming: For better user experience with long outputs
• Connection pooling: Reuse boto3 clients
• Regional deployment: Use closest region to reduce latency
• Provisioned throughput: For consistent high-volume workloads

Security

• Least privilege IAM: Only grant needed model access
• VPC endpoints: Keep traffic private
• Guardrails: Implement content filtering
• Audit with CloudTrail: Track model invocations

IAM Permissions

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "bedrock:InvokeModel",
        "bedrock:InvokeModelWithResponseStream"
      ],
      "Resource": [
        "arn:aws:bedrock:us-east-1::foundation-model/anthropic.claude-3-sonnet-20240229-v1:0",
        "arn:aws:bedrock:us-east-1::foundation-model/amazon.titan-embed-text-v2:0"
      ]
    }
  ]
}

Troubleshooting

AccessDeniedException

Causes:

• Model access not enabled in console
• IAM policy missing bedrock:InvokeModel
• Wrong model ID or region

Debug:

# Check model access status
aws bedrock list-foundation-models \
  --query 'modelSummaries[?modelId==`anthropic.claude-3-sonnet-20240229-v1:0`]'

ThrottlingException

Causes:

• Exceeded on-demand quota
• Too many concurrent requests

Solutions:

• Request quota increase
• Implement exponential backoff
• Consider provisioned throughput

ValidationException

Common issues:

• Invalid model ID
• Malformed request body
• max_tokens exceeds model limit

Retry with Exponential Backoff

import time
from botocore.exceptions import ClientError

def invoke_with_retry(bedrock, body, max_retries=3):
    for attempt in range(max_retries):
        try:
            return bedrock.invoke_model(
                modelId='anthropic.claude-3-sonnet-20240229-v1:0',
                body=json.dumps(body)
            )
        except ClientError as e:
            if e.response['Error']['Code'] in ('ThrottlingException', 'ModelNotReadyException'):
                time.sleep(2 ** attempt)
            else:
                raise
    raise Exception('Max retries exceeded')

References

• Bedrock User Guide
• Bedrock API Reference
• Bedrock Runtime API
• Model Parameters
• Bedrock Pricing