OS Chat

OS Chat is an open-source AI chatbot application built with TanStack Start, Convex, and the Vercel AI SDK.
It supports multiple AI providers, real-time messaging, and chat organization.

Features

AI Model Support

• Google Gemini (2.0 Flash, 2.5 Flash, 2.5 Pro)
• OpenAI (GPT-4o, GPT-4o Mini, GPT-5, o3)
• Anthropic (Claude 3.5 Sonnet, 3.7 Sonnet, Sonnet 4)
• OpenRouter (100+ models via a single API)
• BYOK (Bring your own API keys)

Chat Management

• Real-time streaming responses
• Organize chats with folders
• Pin or branch conversations
• Share chats with others
• Edit or regenerate messages

UI

• Light and dark themes
• Code syntax highlighting (Shiki)
• Math rendering (KaTeX)
• Markdown support

Authentication

• GitHub OAuth (via Better Auth)

Tech Stack

Frontend

• TanStack Start (full-stack React framework)
• TanStack Query (server state management)
• ShadCN (UI components)
• Zustand (client state management)

Backend

• Convex (real-time database and backend)
• Better Auth (authentication library)
• Vercel AI SDK (AI model integration)
• Zod (schema validation)

AI Providers

• OpenAI (GPT models)
• Anthropic (Claude models)
• Google (Gemini models)
• OpenRouter (unified AI API)
• BYOK support for all major providers

Development

• Ultracite (code formatting and linting)
• Biome (fast linter and formatter)

Installation

Prerequisites

• Node.js 18+
• pnpm (recommended) or npm
• Git

Setup

1. Clone the repository:

   git clone https://github.com/ankitk26/os-chat.git
   cd os-chat

2. Install dependencies:

   pnpm install

3. Set up Convex:

   npx convex dev

   Follow the prompts to create a new Convex project and get your deployment URL.

4. Create a .env.local file in the root directory:

   # Convex (Required)
   VITE_CONVEX_URL="https://your_project_url.convex.cloud"
   
   # BetterAuth (Required for authentication)
   BETTER_AUTH_SECRET="your_better_auth_secret"
   BETTER_AUTH_URL="http://localhost:3000"
   
   # GitHub OAuth (Required for authentication)
   GITHUB_CLIENT_ID="your_github_client_id"
   GITHUB_CLIENT_SECRET="your_github_client_secret"
   
   # Optional: Server-side API keys for default access
   OPENROUTER_API_KEY="your_openrouter_api_key"
   OPENAI_API_KEY="your_openai_api_key"
   ANTHROPIC_API_KEY="your_anthropic_api_key"
   GOOGLE_GENERATIVE_AI_API_KEY="your_google_api_key"
   XAI_API_KEY="your_xai_api_key"

5. Run the development server:

   pnpm dev

6. Open your browser and navigate to http://localhost:3000

Tech Stack

Frontend

• TanStack Start - Full-stack React framework
• TanStack Query - Server state management
• ShadCN - UI components
• Zustand - Client state management

Backend

• Convex - Real-time database and backend
• Better Auth - Authentication system
• Vercel AI SDK - AI model integration
• Zod - Schema validation

AI Providers

• OpenAI - GPT models
• Anthropic - Claude models
• Google - Gemini models
• OpenRouter - Unified AI API
• BYOK Support - All major providers

Development

• Ultracite - Code formatting and linting
• Biome - Fast linter and formatter

Usage

1. Sign in with your GitHub account
2. Configure API keys in settings (optional - you can use OpenRouter for all models)
3. Start chatting with any available AI model
4. Organize chats by creating folders and pinning important conversations

License

This project is open-source and licensed under the MIT License.

Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Documentation: Check the code comments and type definitions