A cross-platform AI chat application built with Flutter, supporting multiple AI providers and real-time streaming conversations.
| Chat Interface | Navigation Drawer |
|---|---|
![]() |
![]() |
- Multi-Provider Support: OpenAI, DeepSeek, Anthropic, Google, Ollama, Phind, and more
- AI Assistants: Create personalized AI assistants with custom prompts and parameters
- Real-time Streaming: Support for both streaming and non-streaming responses
- Reasoning Models: Support for advanced reasoning models like OpenAI o1 and DeepSeek R1
- Model Favorites: Bookmark frequently used models for quick access
- Chat History: Complete conversation management and search functionality
- Cross-Platform: Runs on Android, iOS, Windows, macOS, Linux, and Web
- Material Design 3: Modern UI with dynamic color support
- Flutter 3.8.0+ - Cross-platform UI framework
- Dart 3.5.0+ - Programming language
- Material Design 3 - Modern design system with dynamic colors
- Riverpod 2.6+ - State management
- Drift 2.16+ - SQLite database ORM
- llm_dart - Open-source AI interface library supporting multiple LLM providers (MIT License)
- Flutter SDK 3.8.0 or higher
- Dart SDK 3.5.0 or higher
- Clone the repository:
git clone https://github.com/Latias94/yumcha.git
cd yumcha- Install dependencies:
flutter pub get- Run the application:
flutter run- Add AI Providers: Configure your AI service providers with API keys
- Create Assistants: Set up AI assistants with custom prompts and parameters
- Start Chatting: Create conversations and switch between different models
This project is licensed under the AGPL v3 License.
- llm_dart - A standalone open-source AI interface library supporting multiple LLM providers (MIT License). Originally developed as part of YumCha, now independently published on pub.dev.
The UI design of this application is primarily inspired by rikkahub - A native Android LLM chat client that supports switching between different providers for conversations. We appreciate their excellent design and user experience contributions to the AI chat client community.

