Bedrock Chat

A modern web-based chat interface for AWS Bedrock foundation models. Chat with Claude, Llama, Mistral, Nova, and many other AI models through a clean, responsive UI.

Features

• Multi-model support - Access 50+ foundation models from Anthropic, Amazon, Meta, Mistral, Cohere, AI21, DeepSeek, Google, OpenAI, Qwen, NVIDIA, and more
• Streaming responses - Real-time token streaming for faster interactions
• Conversation management - Create, rename, and delete conversations with persistent history
• Model discovery - Automatically fetches available models from your Bedrock account
• Markdown support - Rich formatting with code syntax highlighting
• Secure - Credentials stay in your browser, never sent to any server except AWS

Prerequisites

• Node.js 18+
• An AWS account with Bedrock access
• IAM credentials with appropriate permissions

Quick Start

1. Clone the repository

   git clone https://github.com/yourusername/bedrock-chat.git
   cd bedrock-chat

2. Install dependencies

   npm install

3. Start the development server

   npm run dev

4. Open your browser Navigate to http://localhost:5173

5. Enter your AWS credentials

   • Access Key ID
   • Secret Access Key
   • Region (e.g., us-east-1)
   • Session Token (optional, for temporary credentials)

6. Start chatting! Select a model and begin your conversation.

AWS Setup

Required IAM Permissions

Your IAM user needs the following permissions:

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

Model Access

Before using any model, you must request access in the AWS Bedrock console:

1. Go to AWS Bedrock Console
2. Navigate to Model access in the left sidebar
3. Click Modify model access
4. Select the models you want to use
5. Save your changes

Architecture

┌─────────────────┐     ┌─────────────────┐     ┌─────────────────┐
│   React Client  │────▶│  Express Server │────▶│  AWS Bedrock    │
│   (Port 5173)   │     │   (Port 3001)   │     │     API         │
└─────────────────┘     └─────────────────┘     └─────────────────┘

• Frontend: React + TypeScript + Vite
• Backend: Express.js proxy server
• AWS SDK: @aws-sdk/client-bedrock and @aws-sdk/client-bedrock-runtime

Available Scripts

Command	Description
npm run dev	Start both client and server in development mode
npm run dev:client	Start only the Vite dev server
npm run dev:server	Start only the Express server with hot reload
npm run build	Build the client for production
npm run preview	Preview the production build

Supported Models

The app supports all Bedrock foundation models including:

• Anthropic: Claude 3 Haiku, Claude 3.5 Haiku, Claude Sonnet 4/4.5/4.6, Claude Opus 4.1/4.5/4.6
• Amazon: Nova Micro/Lite/Pro/Premier, Nova 2 Lite, Titan Text
• Meta: Llama 3, 3.1, 3.2, 3.3, Llama 4 Scout/Maverick
• Mistral: Mistral 7B, Mixtral 8x7B, Mistral Large, Ministral, Magistral, Devstral, Pixtral
• Cohere: Command R, Command R+
• AI21 Labs: Jamba 1.5 Large/Mini
• DeepSeek: DeepSeek-R1, DeepSeek V3.x
• Google: Gemma 3 variants
• OpenAI: GPT OSS models
• Qwen: Qwen3 variants
• NVIDIA: Nemotron models
• Moonshot: Kimi K2 variants
• Writer: Palmyra X4/X5
• MiniMax: MiniMax M2/M2.1
• Z.AI: GLM 4.7 variants

Security Notes

• Credentials are never stored on any server - They remain in your browser's localStorage (if you enable "Remember credentials") or in memory only
• No backend credential storage - The Express server only proxies requests to AWS
• No telemetry or analytics - Your conversations stay private
• Open source - Audit the code yourself

Tech Stack

• React 18 - UI framework
• TypeScript - Type safety
• Vite - Build tool
• Express - Backend server
• AWS SDK for JavaScript v3 - AWS integration
• react-markdown - Markdown rendering
• highlight.js - Code syntax highlighting

License

MIT License - feel free to use this project for any purpose.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Acknowledgments

Built with ❤️ using AWS Bedrock and React.