Before you begin, you'll need:
- A Stream API Key (Register for free)
- OpenAI API credentials (Sign up)
- Anthropic API credentials (Sign up)
This repository showcases official Stream sample projects demonstrating Generative AI integration with Chat API and UI Kits.
Frontend SDKs:
- React for web apps
- React Native for mobile apps
- Jetpack Compose for Android
- Swift UI for iOS
- Flutter for cross-platform development
Backend Examples:
- NodeJS implementation
- Python AI Chat implementation
You can use our pre-built OpenAI/Anthropic integrations or extend the Agent class to work with any external LLM provider.
The repository implements a modular architecture that connects Stream's Chat functionality with various LLM providers:
Key features:
- Pre-built integrations with OpenAI and Anthropic
- Extensible Agent class for custom LLM provider integration
- Built-in UI components that handle:
- Loading states
- Code block rendering
- Markdown formatting
- Table displays
- Image rendering
- Clone the repo
- Configure environment variables:
cp .env.example .env- Add the following credentials to your
.env:
- Follow the setup instructions in the nodejs-ai-assistant directory
- Start reading our comprehensive backend integration guide written by Stream engineers explaining the setup and integration with Stream Chat.
Select your preferred SDK to get started.
- How to Build an AI Assistant with React
- How to Build an AI Assistant with React Native
- How to Build an AI Assistant for Android Using Compose
- How to Build an AI Assistant for iOS Using Swift UI
- How to Build an AI Assistant with Flutter
These guides explore more of how AI can be used across our SDKs and products.
- How to Build an AI Chat Android App With Google’s Generative AI
- How to Build an LLM-powered Chatbot For Your Documentation
- How to Build an AI bot Using Stream Chat, HuggingFace, and DialogGPT
- How to Build AI-Powered Chatbot Apps for Android Using Firebase
- How to Add RAG-Based AI to Team Chat With Stream
- How to Build an Agentic RAG System With OpenAI, LanceDB, and Phidata
- How to Chat With Any Book Using Pinecone, OpenAI, and Stream
We welcome contributions that improve functionality or fix issues:
- Fork the repository
- Create a feature branch
- Implement your changes
- Add tests if applicable
- Submit a PR with a clear description/changes
We encourage you to try the tutorials and submit new sample apps 🥳
- How can I submit a sample app?
- Open a pr with a proper description and we'll review it as soon as possible.
- Open an issue with:
- Clear steps to reproduce
- Add screenshots/videos if possible
- Expected behavior
- Actual behavior
- Platform/environment details