Personal Jarvis
Personal Jarvis is an open-source, voice-controlled AI assistant that can operate your computer, run shell commands, use coding agents, and connect to MCP services. It solves the problem of needing multiple tools and manual control by letting you speak commands to get things done. It is for developers and power users who want a customizable, local-first AI assistant.
🎬 AI Voice Assistant – Personal Jarvis · Personal Jarvis
✨ Key features
- Voice-controlled operation of computer, shell, and coding agents
- Runs fully local with no cloud account required
- Connects to any MCP-compatible service
- Remembers information across sessions via Knowledge Wiki
- Switches between local and hosted AI models per tier
- Includes Agentic IDE with terminal and chat
🎯 Use cases
- Research topics and generate reports via voice
- Control desktop applications and settings by voice
- Run coding tasks in isolated workspaces with AI agents
- Make outbound phone calls via Twilio integration
- Dictate into any application using voice
📦 Installation
🧰 Requirements: Python 3.11+ and Git; optional API keys for hosted models, Node.js 18+ for some coding agents, libportaudio on Linux for local audio.
Install it in one command
Windows (PowerShell)
irm https://raw.githubusercontent.com/PersonalJarvis/PersonalJarvis/main/install/install.ps1 | iex
macOS and Linux
curl -fsSL https://raw.githubusercontent.com/PersonalJarvis/PersonalJarvis/main/install/install.sh | bash
Python 3.11+ and Git, nothing else. The installer asks nothing in the terminal, and the app walks you through language, wake word, and keys once.
🚀 Usage
jarvis # full desktop: window + voice + Orb overlay
jarvis serve # headless server: API + WebSocket + browser UI, no local audio needed
⚠️ Good to know
Outbound phone calls require Twilio (hosted service) and coding-agent workers run on external CLIs or API keys; local realtime model is experimental and needs ~12 GB GPU memory.
❓ FAQ
Can I run Personal Jarvis without any cloud account?
Yes, every layer has a keyless local option, including local speech recognition, text-to-speech, and a self-hosted realtime server, so the entire assistant can run on your own hardware.
What are the system requirements?
You need Python 3.11+ and Git. For local realtime model, about 12 GB of GPU or unified memory is recommended. A GPU is optional for faster speech processing.
How do I update Personal Jarvis?
Re-running the same one-liner install command updates in place.
Does it cost money?
Personal Jarvis is open-source (Apache 2.0) and free to use. You only pay for the AI models you choose to use, billed by the provider directly.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.