Develop Generative AI Applications Using Amazon Bedrock with AWS SDK for Python (Boto3)

Introduction

Amazon Bedrock is a serverless platform that provides flawless access to important foundation models (FMs) from leading AI originators like AI21 Labs, Anthropic, Cohere, Meta, Mistral AI, Stability AI, and Amazon. With a single API, Amazon Bedrock empowers inventors to experiment, customize, and emplace these models for different use cases.

It supports advanced customization ways like fine-tuning and retrieving augmented generation (RAG) while icing security, sequestration, and responsible AI practices.

By barring structure operation, Amazon Bedrock allows integration with familiar AWS services for structure and planting generative AI operations efficiently.

Pioneers in Cloud Consulting & Migration Services

Reduced infrastructural costs
Accelerated application deployment

Get Started

Solution Overview

The solution uses an AWS SDK for Python script with features that bring Anthropic Claude 3 Sonnet on Amazon Bedrock. Using this FM generates an affair using a prompt as input. The following illustration illustrates the result armature.

Prerequisites

Before using the Amazon Bedrock API, ensure you have the following prerequisites in place:

An active AWS account with permission to access services like Amazon Bedrock.
The AWS Command Line Interface (CLI) was installed and configured.
An AWS Identity and Access Management (IAM) user was created with the required permissions to interact with the Amazon Bedrock API.
AWS IAM user’s access and secret keys are used to configure the AWS CLI and manage permissions.
Authorization to access foundation models (FMs) available on Amazon Bedrock.
The latest version of the Boto3 library is installed.
Python version 3.8 or higher, properly set up within your integrated development environment (IDE).

Deploy the solution

Import the required libraries:

import boto3
import json

1 2	import boto3 import json

2. Configure the Boto3 client to interact with the Amazon Bedrock runtime and define the desired AWS Region:

# Set up the Amazon Bedrock client
bedrock_client = boto3.client(
    	service_name="bedrock-runtime",
    region_name="us-east-1"
)

# Set up the Amazon Bedrock client

bedrock_client = boto3.client(

service_name="bedrock-runtime",

region_name="us-east-1"

)

3. Specify the model to invoke by providing its model ID. For instance, this example utilizes the Anthropic Claude 3 Sonnet, available on Amazon Bedrock.

# Define the model ID
model_id = "anthropic.claude-3-sonnet-20240229-v1:0"

1 2	# Define the model ID model_id = "anthropic.claude-3-sonnet-20240229-v1:0"

4. Provide a prompt, the input message used to interact with the foundation model during invocation.

# Prepare the input prompt.
   prompt = "Hello, how are you?"

1 2	# Prepare the input prompt. prompt = "Hello, how are you?"

Prompt engineering techniques can significantly enhance the performance of foundation models (FMs) and improve the quality of their outputs.

Before invoking a model on Amazon Bedrock, you must define a payload. This payload is a set of instructions and contextual information that guides the model generation process. The payload structure depends on the model you are using.

For example, when working with Anthropic Claude 3 Sonnet on Amazon Bedrock, the payload acts as a blueprint, providing the necessary context and parameters for the model to generate text based on your prompt. Key components of this payload include:

anthropic_version: Specifies the version of Amazon Bedrock being utilized.
max_tokens: Sets the maximum number of tokens the model can generate in a single response. Tokens are basic units of text, including words, subwords, or punctuation, processed by large language models (LLMs).
temperature: Controls the randomness of the output. Higher values encourage creativity and diverse responses, while lower values produce more predictable and consistent results.
top_k: Determines the number of top candidate words considered at each step of the text generation process.
top_p: Modifies the sampling probability distribution, prioritizing frequent words with higher values and promoting diversity with lower values.
messages: Represents an array of messages for the model to interpret and process.
role: Indicates the sender’s role in the message, such as “user” for the input prompt.
content: Contains the actual text of the prompt, encapsulated as a “text” type object.

5. Define the payload as follows:

payload = {
    "anthropic_version": "bedrock-2023-05-31",
    "max_tokens": 2048,
    "temperature": 0.9,
    "top_k": 250,
       "top_p": 1,
    "messages": [
        {
            "role": "user",
            "content": [
                {
                    "type": "text",
                    "text": prompt
               }
            ]
        }
    ]
}

payload = {

"anthropic_version": "bedrock-2023-05-31",

"max_tokens": 2048,

"temperature": 0.9,

"top_k": 250,

"top_p": 1,

"messages": [

{

"role": "user",

"content": [

{

"type": "text",

"text": prompt

}

]

}

]

}

6. After configuring the parameters and selecting the desired foundation model (FM), a request can be sent to Amazon Bedrock. This request includes the chosen FM and the defined payload, which contains the necessary instructions for the model to process.

# Invoke the Amazon Bedrock model
response = bedrock_client.invoke_model(
    modelId=model_id,
    body=json.dumps(payload)
)

# Invoke the Amazon Bedrock model

response = bedrock_client.invoke_model(

modelId=model_id,

body=json.dumps(payload)

)

7. Once the request is processed, the generated text result from Amazon Bedrock can be displayed.

# Process the response
result = json.loads(response["body"].read())
generated_text = "".join([output["text"] for output in result["content"]])
print(f"Response: {generated_text}")

# Process the response

result = json.loads(response["body"].read())

generated_text = "".join([output["text"] for output in result["content"]])

print(f"Response: {generated_text}")

Invoke the Model

Invoking the model with the prompt “Hello, how are you?” will produce the result in the screenshot below.

Conclusion

In conclusion, how to utilize Amazon Bedrock advanced foundation models (FMs) with the AWS SDK for Python (Boto3), walking through the steps of invoking models and processing the generated outputs, shown how these models can be applied to various use cases, such as creative content generation, code assistance, data summarization, and conversational AI. As generative AI technologies evolve, exploring and comparing different models allows developers to select the best fit for their needs. Amazon Bedrock provides a robust platform for building AI-driven applications, and with continued experimentation, these models can be used to drive innovation and transform industries.

Drop a query if you have any questions regarding Amazon Bedrock and we will get back to you quickly.

Making IT Networks Enterprise-ready – Cloud Management Services

Accelerated cloud migration
End-to-end view of the cloud environment

Get Started

About CloudThat

CloudThat is a leading provider of Cloud Training and Consulting services with a global presence in India, the USA, Asia, Europe, and Africa. Specializing in AWS, Microsoft Azure, GCP, VMware, Databricks, and more, the company serves mid-market and enterprise clients, offering comprehensive expertise in Cloud Migration, Data Platforms, DevOps, IoT, AI/ML, and more.

CloudThat is the first Indian Company to win the prestigious Microsoft Partner 2024 Award and is recognized as a top-tier partner with AWS and Microsoft, including the prestigious ‘Think Big’ partner award from AWS and the Microsoft Superstars FY 2023 award in Asia & India. Having trained 650k+ professionals in 500+ cloud certifications and completed 300+ consulting projects globally, CloudThat is an official AWS Advanced Consulting Partner, Microsoft Gold Partner, AWS Training Partner, AWS Migration Partner, AWS Data and Analytics Partner, AWS DevOps Competency Partner, AWS GenAI Competency Partner, Amazon QuickSight Service Delivery Partner, Amazon EKS Service Delivery Partner, AWS Microsoft Workload Partners, Amazon EC2 Service Delivery Partner, Amazon ECS Service Delivery Partner, AWS Glue Service Delivery Partner, Amazon Redshift Service Delivery Partner, AWS Control Tower Service Delivery Partner, AWS WAF Service Delivery Partner, Amazon CloudFront and many more.

To get started, go through our Consultancy page and Managed Services Package, CloudThat’s offerings.

FAQs

1. What are the prerequisites to using Amazon Bedrock with Boto3?

ANS: – You need an AWS account, AWS CLI configured, an AWS IAM user with permissions, the latest Boto3 library, and Python 3.8+ in your development environment.

2. How do you configure the Boto3 client for Amazon Bedrock?

ANS: – Configure the client by specifying the service name as “bedrock-runtime” and the desired region, like this:

bedrock_client = boto3.client(service_name=”bedrock-runtime”, region_name=”us-east-1″)

WRITTEN BY Aayushi Khandelwal

Aayushi, a dedicated Research Associate pursuing a Bachelor's degree in Computer Science, is passionate about technology and cloud computing. Her fascination with cloud technology led her to a career in AWS Consulting, where she finds satisfaction in helping clients overcome challenges and optimize their cloud infrastructure. Committed to continuous learning, Aayushi stays updated with evolving AWS technologies, aiming to impact the field significantly and contribute to the success of businesses leveraging AWS services.