OpenJarvis: a local-first personal AI is now available to run with Ollama

OpenJarvis v1.0 is now available, according to its publisher. The open-source framework is for building personal AI agents that run on the user's own hardware, with Ollama support built in.

It is built by Stanford's Hazy Research and Scaling Intelligence labs as part of their "Intelligence Per Watt" research into efficient local AI. According to the release, local models can already handle most day-to-day chat and reasoning, yet most personal AI still sends every request to the cloud. OpenJarvis makes local-first the default: models run locally and the cloud is optional. Energy, cost, and latency are tracked alongside accuracy.

Installation begins with Ollama, available for macOS, Windows, or Linux. On macOS or Linux, an install script auto-detects an existing Ollama installation; on Windows, the command runs inside WSL2 or users can install a desktop app. Running `jarvis` starts it.

The install script sets up a starter model, and users can pull other models through Ollama or set a default model in a config file with the preferred engine set to ollama.

OpenJarvis ships with ready-to-run presets, each bundling an agent with the engines and tools it needs. A morning briefing preset generates a briefing from calendar, email, and the day's news. A deep-research preset answers questions across the web and local documents with citations. A code-assistant preset writes and runs Python on the user's machine. Links to GitHub, documentation, and v1.0.0 release notes were provided.