The hard part of a supervised agent is how often to ask the human: too often and they stop reading, too rarely and something slips through. Jev gives a number from 0 to 1: above 0.9 the agent proceeds, below 0.1 it stops, and in between it asks.
The hard part of a supervised agent: asking the human too often and they stop reading; too rarely and something slips through. Jev gives you a number instead of a sentence. Say yes above 0.9 and the agent goes ahead. Say no below 0.1 and it stops. Anything in between, it asks
Gregor Zunic showcases an open-source browser agent combining Browser Use and Jev, finding flights in 7 seconds for $0.0039, using a new action space per step, DOM state space, and small LLM fallback for typing.
Scott Williams predicts custom Jev-style models (parallel constrained decoding) will make existing company agent systems more token efficient, creating millionaires.
The user praises the JEV model for its extremely fast reproduction of each movement and command, believing bots will become smarter as a result.
