Project Overview
A modern AI chat application that provides a seamless conversational experience with support for multiple language models including GPT-4, Claude, and open-source alternatives.
Key Features
- Streaming Responses: Real-time token-by-token response rendering
- Multi-Model: Switch between OpenAI, Anthropic, and local models
- Rich Rendering: Full markdown, code highlighting, and LaTeX support
- Chat History: Persistent conversations with search and organization
Technical Highlights
- Server-Sent Events for streaming with backpressure handling
- Custom markdown renderer with syntax highlighting (Shiki)
- Optimistic UI updates for instant feedback
- IndexedDB for offline-first chat persistence