kitio.aiAdded Oct 30, 2026
OpenJarvis
FreeOpen-source framework for personal AI agents that run on your own device
What is OpenJarvis?
OpenJarvis is an Apache-licensed framework from Stanford for building personal AI agents that run locally by default and call cloud models only when needed. It suits developers and researchers who want private, on-device assistants for briefings, research and coding.
A one-line installer sets up Python, Ollama and a starter model, then the jarvis command opens a chat. Presets configure agents for a spoken morning digest from email and calendar, multi-hop research with citations, a code assistant with shell access, and scheduled monitoring with memory.
Agents can load skills from Hermes Agent, OpenClaw or any GitHub repo, and a learning loop tunes models and skills from local traces. Its benchmarks track energy, latency and dollar cost next to accuracy. It comes from Hazy Research and the Scaling Intelligence Lab at Stanford, with desktop builds for macOS, Windows and Linux.
Features
- Local-first agent runtime
- Morning digest preset
- Deep research agent
- Skill imports
- Energy and cost benchmarks
Tags
Releases
+ Submit new release- Listed
OpenJarvis was added to the directory.
Organization
Stanford Hazy Research
🇺🇸 United States
Listed by kitio.ai · unverified owner
Pricing
- Pricing model
- Free
- Paid options from
- 100% free
OpenJarvis plans
Open source
Free
- Apache 2.0 license
- Runs on your own hardware
- Optional cloud model engines
Prices in USD, checked Oct 30, 2026. Plans change often, so confirm on the OpenJarvis site.Official pricing
Sponsored
Your AI tool here
A Featured launch puts your tool in this spot and on the home page for a week.
Top OpenJarvis alternatives
View allAgents & automation
Open-source personal agent with editable Markdown memory
Agents & automation
Source-available workflow automation you can self-host
Agents & automation
Low-level framework for building stateful, controllable agents
Compare OpenJarvis
If you liked OpenJarvis
Run open models locally with one command, plus optional cloud models
Freemium · $20/moModels & APIsMade in United StatesOpen-source, model-agnostic coding agent for the terminal
Freemium · $10/moCodingMade in United StatesFree open-source desktop client for dozens of AI providers
FreeChat assistantsMade in ChinaApple's array framework for running and training models on Apple Silicon
FreeModels & APIsMade in United States
Pros and cons
Pros
- Free and Apache 2.0 licensed
- Runs on local models by default
- Ready-made presets for common agents
Cons
- Browser GUI from source needs Node.js 22+
- Larger local models need strong hardware
- Young project, so features change often
Prompts & results
No prompts yet. Add your own prompt and output to help others use this tool.
Share a prompt
Reviews
Q&A
What is OpenJarvis?
OpenJarvis is an Apache-licensed framework from Stanford for building personal AI agents that run locally by default and call cloud models only when needed. It suits developers and researchers who want private, on-device assistants for briefings, research and coding.
Does OpenJarvis need a cloud API key?
No. It runs local models through Ollama by default and only uses cloud models if you configure them.
How do I install OpenJarvis?
Run the one-line install script for macOS, Linux, WSL2 or Windows, or download a desktop build from the GitHub releases page.
How much does OpenJarvis cost?
OpenJarvis is 100% free.
Is there a free version of OpenJarvis?
Yes. OpenJarvis is completely free to use.
What are the main features of OpenJarvis?
Local-first agent runtime, Morning digest preset, Deep research agent, Skill imports, Energy and cost benchmarks.
Who makes OpenJarvis?
OpenJarvis is made by Stanford Hazy Research.
What are the best alternatives to OpenJarvis?
Popular alternatives include OpenClaw, Hermes Agent, n8n.