This repository is dedicated to exploring and experimenting with Vercel's AI SDK, showcasing various implementations and use cases for building AI-powered applications.
The AI SDK from Vercel provides a comprehensive toolkit for building AI applications with streaming, tool calling, and multi-modal capabilities. This repository serves as a collection of experiments, examples, and projects that demonstrate the SDK's powerful features.
Directory: chat-with-mcp/
A modern chat application featuring multi-provider AI support (Anthropic Claude and OpenAI GPT) with Model Context Protocol integration for extensible tools. Includes real-time streaming, rich message rendering with multiple part types, and visual tool execution tracking. Supports file operations, web search, and persistent memory through MCP servers. Built with Next.js 15, TypeScript, and shadcn/ui for a polished user experience.
Planned experiments and explorations:
- Multi-modal Chat: Image, audio, and video processing capabilities
- RAG Implementation: Retrieval-Augmented Generation with vector databases
- Agent Workflows: Multi-step agent interactions and decision trees
- Custom Model Integration: Integration with local and custom AI models
- Real-time Collaboration: Multi-user AI-assisted collaboration tools
- Voice Interface: Speech-to-text and text-to-speech integration
- Code Generation: AI-powered code assistance and generation tools
Contributions are welcome! Whether you want to add new experiments, improve existing implementations, fix bugs, or share interesting use cases, please feel free to open issues or submit pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.
Happy experimenting! 🧪✨