Post
Ask Question

Conversation with document using Amazon Bedrock and .NET

Introduction

In this article, you'll learn how to send a document as part of a message using the Amazon Bedrock Converse API with the Anthropic Claude 3 Sonnet model and request that the model describe the contents of the document.

Prerequisites

  1. Create an AWS account and log in. Ensure the IAM user you use has sufficient permissions to make necessary AWS service calls and manage AWS resources.
  2. Download and install the AWS Command Line Interface (CLI).
  3. Configure the AWS CLI.
  4. Download and install Visual Studio or Visual Studio Code.
  5. Download and install .NET 8.0 SDK
  6. Access to Amazon Bedrock foundation model

Tools

  1. Visual Studio 2022

Steps Involved

Perform the following steps to create a .NET console application in Visual Studio 2022 to send a document as part of a message and request that the model describe the contents of the document.

  1. Open Visual Studio 2022.
  2. Click File -> New -> Project.
  3. Select the Console App template. Click Next.
  4. Enter the project name and click Next.
  5. Select the .NET 8.0 framework. Click Create.
  6. Add the following NuGet packages. 
    AWSSDK.BedrockRuntime
  7. Open Program.cs and replace the code with the following. 
    using Amazon;
using Amazon.BedrockRuntime;
using Amazon.BedrockRuntime.Model;

namespace AmazonBedrockConverseApp
{
    internal class Program
    {
        // Constants for the model ID, input text, and input document path
        private const string ModelId = "anthropic.claude-3-sonnet-20240229-v1:0";
        private const string InputText = "What's in this document?";
        private const string InputDocumentPath = "C:\\Users\\admin\\source\\repos\\AmazonBedrockConverseApp\\India.txt";

        static async Task Main(string[] args)
        {
            // Configure the Amazon Bedrock Runtime client
            var config = new AmazonBedrockRuntimeConfig
            {
                RegionEndpoint = RegionEndpoint.USEast1 // Use your region
            };

            using var bedrockClient = new AmazonBedrockRuntimeClient(config);

            try
            {
                // Generate a message and print the response
                var response = await GenerateMessageAsync(bedrockClient, ModelId, InputText, InputDocumentPath);
                PrintResponse(response);
            }
            catch (AmazonBedrockRuntimeException ex)
            {
                // Handle any errors that occur during the conversation
                Console.WriteLine($"Error: {ex.Message}");
            }
        }

        /// <summary>
        /// Generates a message using the Bedrock Runtime client.
        /// </summary>
        /// <param name="bedrockClient">The Bedrock Runtime client.</param>
        /// <param name="modelId">The model ID to use.</param>
        /// <param name="inputText">The input text for the message.</param>
        /// <param name="inputDocumentPath">The path to the input document file.</param>
        /// <returns>The response from the Bedrock Runtime.</returns>
        private static async Task<ConverseResponse> GenerateMessageAsync(IAmazonBedrockRuntime bedrockClient, string modelId, string inputText, string inputDocumentPath)
        {
            // Read the document file as a byte array and convert to a MemoryStream
            byte[] documentBytes = await File.ReadAllBytesAsync(inputDocumentPath);
            using var documentStream = new MemoryStream(documentBytes);

            // Create the message with text and document content
            var message = new Message
            {
                Role = "user",
                Content = new List<ContentBlock>
                {
                    new()
                    {
                        Text = inputText
                    },
                    new()
                    {
                        Document = new DocumentBlock
                        {
                            Name = "MyDocument",
                            Format = DocumentFormat.Txt,
                            Source = new DocumentSource { Bytes = documentStream }
                        }
                    }
                }
            };

            // Create the Converse request
            var request = new ConverseRequest
            {
                ModelId = modelId,
                Messages = new List<Message> { message }
            };

            // Send the Converse request and return the response
            return await bedrockClient.ConverseAsync(request);
        }

        /// <summary>
        /// Prints the response from the Bedrock Runtime to the console.
        /// </summary>
        /// <param name="response">The response from the Bedrock Runtime.</param>
        private static void PrintResponse(ConverseResponse response)
        {
            Console.WriteLine("Role: " + response.Output.Message.Role);

            // Print each content block in the response
            foreach (var content in response.Output.Message.Content)
            {
                if (content.Text != null)
                {
                    Console.WriteLine("Text: " + content.Text);
                }
            }

            // Print the usage statistics
            Console.WriteLine("Input tokens: " + response.Usage.InputTokens);
            Console.WriteLine("Output tokens: " + response.Usage.OutputTokens);
            Console.WriteLine("Total tokens: " + response.Usage.TotalTokens);
            Console.WriteLine("Stop reason: " + response.StopReason);
        }
    }
}
  8. Press F5 to run the application. You should see the output from the model describing the contents of the document.
    Result

Summary

This article describes how to create a .NET console application to send a document as part of a message using the Amazon Bedrock Converse API with the Anthropic Claude 3 Sonnet model and requests that the model describe the contents of the document.in Visual Studio 2022.


Similar Articles