Skip to content

Every AI model, one blazing fast workspace - see how they think, schedule what you need

License

Notifications You must be signed in to change notification settings

ajanraj/OpenChat

Repository files navigation

OS Chat

Visit OS Chat

OS Chat is a free, open-source AI personal assistant that combines 50+ language models with powerful automation capabilities. Deploy background agents, connect your favorite services (Gmail, Calendar, Notion, GitHub), and get things done through natural conversation.

OS Chat — Open-source AI chat app (oschat.ai)

🚀 Why OS Chat?

  • 🔓 Open Source T3 Chat Alternative - Get all the speed and multi-model access of T3 Chat, but open source, self-hostable, and enhanced with background agents and service integrations
  • 🤖 Your AI Personal Assistant - More than chat - deploy background agents, automate workflows, and manage your digital life
  • 🔗 Connect Everything - Direct integration with Gmail, Calendar, Notion, GitHub, Slack, and 10+ more services via Composio
  • ⏰ Smart Automation - Deploy background agents with email notifications - let your assistant work while you sleep
  • 🧠 50+ AI Models - Access the latest from OpenAI, Anthropic, Google, Meta, and more in one interface
  • 🎯 Truly Personal - Customize personality traits, personal context, and preferences for tailored interactions
  • 🔒 Privacy-First - Open source, self-hostable, with encrypted API key storage

✨ Features

⏰ Background Agents & Automation

  • Smart Background Agents - Deploy AI agents to run one-time, daily, or weekly with timezone awareness
  • Email Notifications - Get notified when your background agents complete successfully
  • Automated Workflows - Let your AI assistant handle routine work while you focus on what matters
  • Execution History - Track and monitor all your background agent runs with detailed logs

🔗 Service Connectors & Integrations (via Composio)

  • Gmail Integration - Access, read, and manage your email messages directly through chat
  • Google Calendar - View, create, and schedule calendar events seamlessly
  • Google Drive & Docs - Access files, create documents, and collaborate on projects
  • Notion Workspace - Read and write to your Notion pages and databases
  • GitHub Management - Manage repositories, issues, and pull requests
  • Slack, Linear, X (Twitter) - And 10+ more services

🤖 AI & Models

  • 50+ AI Models - Access OpenAI, Anthropic, Google, Meta, Mistral, DeepSeek, xAI, Moonshot, Z.AI, MiniMax, and Qwen models
  • Multi-modal Support - Text, images, and reasoning across all supported models
  • Image Generation - Create high-quality images with GPT Image 1.5, Nano Banana Pro, Imagen 4, and Flux Schnell
  • Reasoning Models - View AI thinking process with o3, Claude 4, Gemini Thinking, and DeepSeek R1
  • Web Search Integration - Real-time internet search using Exa, Tavily, and Brave APIs

💬 Chat Management

  • Smart Organization - Automatic grouping by Today, Yesterday, Last 7 Days, etc.
  • Pinned Chats - Keep important conversations at the top
  • Chat Branching - Create alternative conversation paths from any assistant message
  • Advanced Search - Full-text search across chat history with content snippets
  • Data Portability - Export/import chat history with full data control

🎨 Interface & Experience

  • Responsive Design - Beautiful interface that works on desktop and mobile
  • Theme System - Beautiful light and dark modes with smooth transitions
  • Keyboard Shortcuts - Quick access with ⌘+K (search), ⌘+Shift+O (new chat), ⌘+B (toggle sidebar)
  • Real-time Streaming - Instant message streaming for immediate responses

🤖 Available Models

💬 Text & Chat Models

  • OpenAI: GPT-5.2, GPT-5.2 Pro, GPT-5.1, GPT-5, GPT-5 Mini/Nano, GPT OSS 20B/120B, o3, o4 Mini
  • Anthropic: Claude 4.5 Opus, Claude 4.5 Sonnet, Claude 4.5 Haiku (with reasoning)
  • Google: Gemini 3 Pro/Flash, Gemini 2.5 Pro/Flash (with thinking)
  • Meta: Llama 4 Maverick, Llama 4 Scout
  • DeepSeek: V3.2, V3.1, R1 (with reasoning variants)
  • xAI: Grok 4.1 Fast, Grok 4, Grok 3
  • Moonshot: Kimi K2, Kimi K2 Thinking
  • Z.AI: GLM 4.7, GLM 4.6, GLM 4.5 (with thinking)
  • MiniMax: M2.1, M2
  • Qwen: Qwen3 Coder, Qwen3 235B

🎨 Image Generation

  • OpenAI: GPT Image 1.5
  • Google: Nano Banana Pro, Nano Banana, Imagen 4, Imagen 4 Ultra
  • Fal: Flux Schnell

🛠️ Built with

🚀 Getting Started

See the Installation Guide for complete setup instructions.

Quick Start

# Clone and install
git clone https://github.com/ajanraj/OpenChat.git
cd OpenChat && bun install

# Set up Convex backend
bunx convex login && bunx convex dev --once

# Configure environment
cp .env.example .env

# Run development server
bun dev

Visit http://localhost:3000 to see OS Chat running locally!

📜 Development Commands

Command Description
bun dev Start development server on port 3000
bun build Build for production
bun test Run tests with Vitest
bun run lint Run Ultracite linter
bun run typecheck Run TypeScript type checking with tsgo
bunx convex dev Run Convex development server

🗺️ Roadmap

  • Projects & Workspaces - Organize your chats into projects and workspaces
  • MCP integration - Model Context Protocol support for enhanced AI capabilities
  • Stream Resuming - Resume interrupted streams using Redis

🤝 Contributing

We welcome contributions! See the Installation Guide to set up your development environment.

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/amazing-feature)
  3. Run bun run lint and bun run format before committing
  4. Push to the branch and open a Pull Request

Areas We'd Love Help With

  • Stream resuming using Redis
  • Performance optimizations
  • MCP integration
  • Additional service connectors

⚠️ Notes

Current Status: Beta Release - OS Chat is actively developed with regular feature updates.

Compatibility: Built with TanStack Start, Vite, and Nitro for modern full-stack React architecture.

Privacy: All data is processed securely with user control over exports, imports, and data management.


Copyright 2025 Ajan Raj. Licensed under the Apache License 2.0.

About

Every AI model, one blazing fast workspace - see how they think, schedule what you need

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Sponsor this project

 

Packages

No packages published

Contributors 5

Languages