Damn, there really is a way to operate a computer faster

Built a Jev-powered computer-use layer for Codex, comparing it against the built-in version on a calendar-event task with fewer pauses.

The original post, translated

Damn, there really is a way to operate a computer faster than Codex's built-in computer use

I tried using Codex + Jev to build an enhanced version of computer use, which I call "Jev Use". It's faster and smoother than the built-in one, yet the token consumption is about the same

Below is a comparison video I made using "adding a Mac calendar event". With the same content, you can clearly see that the Jev version has almost no pauses throughout the whole process 😆

The post above is a machine translation from zh; the untranslated text is in the fold-out.

Quoted post

S

@Saccc_c307.4K views

source

强烈建议大家都亲自试试Jev，能让你的Codex操作速度提高10倍并省下大量token！ 因为Jev非常适合用于判断型工作，能在毫秒内做出正确判断，而且token价格极低。因此非常适合以下几个场景： 1、computer use 将Jev用作操作判断层，指导模型快速操作你的电脑

Engagement when collected

Numbers are a snapshot taken from X when the case was added to the library (schema v1, collected 2026-09-19); they will not match today.