Jev picks the next browser action while a larger model does the thinking
Jev looks at a page and chooses a single next action such as click, type, scroll, buy/sell or hand off, aimed at browser agents, auto QA and trading loops.
Just found Jev. Jev looks at the page, sees the real buttons, and just picks: click this one, type here, scroll, wait, or done. - Click this - Scroll down - Type this - Buy / sell - Retry or stop - Send the job to another agent That’s it. This is useful for: • Browser agents that need to be fast • Auto QA • Trading loops • Agent routing • Games / robots • Anything where the AI has to choose an action over and over. The real trick is splitting thinking from “what do I do next”. Big model does the hard thinking. And the small fast model just picks the next click. Still looking into it, so forgive me for any false claims. Open source here: http://github.com/browser-use/jev-ultrafast