An agent session got stuck at 992k tokens. The user had Jev from TypeSafe AI judge every old tool call while keeping every message verbatim, reducing it to 379k tokens and resuming where it left off.
life saving usecase: one agent session got stuck at 992k tokens. i let @typesafeai jev judge every old tool call, kept every message word for word, and got it to 379k. it picked up where it left off.
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.
