Skip to content

Introduction

A blazingly fast, privacy-first chat interface for AI that works with any LLM provider—cloud or completely offline.

Faster Chat is a self-hosted web application that gives you complete control over your AI conversations. Connect to OpenAI, Anthropic, Groq, Mistral, or run completely offline with Ollama, LM Studio, or llama.cpp.

Privacy-First, Always

  • 🗄️ Conversations stay on your machine - Local-first IndexedDB storage
  • 🔐 Encrypted API key storage - Server-side encryption for provider credentials
  • 🚫 No tracking, no analytics - Your privacy is paramount

Blazingly Fast

  • 3KB Preact runtime - Zero SSR overhead, instant responses
  • 💬 Real-time streaming - Powered by Vercel AI SDK
  • 📱 Responsive design - Works on desktop, tablet, and mobile

Provider-Agnostic

  • 🤖 Multi-provider support - OpenAI, Anthropic, Ollama, Groq, Mistral, custom APIs
  • 🔌 Auto-discover models - Integration with models.dev
  • 🌐 Works completely offline - Run local models with Ollama or LM Studio
  • Real-time streaming responses with the Vercel AI SDK
  • Markdown rendering with syntax highlighting and LaTeX support
  • File attachments with preview and download
  • Conversation history and search
  • Multi-user authentication with role-based access (admin/member/readonly)
  • Provider Hub for managing AI connections and API keys
  • Admin panel for user management (CRUD, password reset, role changes)
  • Auto-discovery of available models from configured providers
  • One-command Docker deployment with optional HTTPS via Caddy
  • Development mode with Bun for fast iteration
  • Production builds optimized for Node.js compatibility

Frontend

Backend

  • Hono - Ultrafast web framework
  • Vercel AI SDK - Multi-provider LLM streaming
  • SQLite - Embedded database for users and configuration
  • Argon2 - Password hashing

Faster Chat is MIT licensed and welcomes contributions:

Built with ❤️ by 1337Hero for developers who value privacy, speed, and control.