Jev handles local computer use at ~90ms per click decision
A local CoreML model segments on-screen elements and on-device OCR reads their labels, so Jev only sees text and picks which button to click, all on-device at about 90ms per decision.
Okay so Jev can actually do computer use really well Without any screenshots, or LLMs and no Pixels leave my mac I dont even read the Dom elements A local CoreML model segments every button and UI element on screen. On-device OCR reads the labels. That text is all Jev gets. It returns a probability across those elements and tells me the best one to click. Then it clicks, re-runs detection, and decides again. In a loop until the goal is done. ~90ms per decision. Faster than any LLM computer use I've tried. Blazing fast computer use, without any latency @typesafeai is building something really interesting
This post quotes another post: “ChatGPT co-inventor releases Jev, a model built for decisions rather than text”