Just came across a very wild new model: **Jev**.
Described Jev as a model that takes state plus custom questions and returns choices with probabilities for branching.
The original post, translated
Just came across a very wild new model: **Jev**. It doesn't chat, doesn't write text, doesn't write code. You throw it a state + a few questions you define yourself, and it only does one thing—**make judgments**. Which queue to pick, how urgent it is, whether to go manual, which button to click next, it directly returns the option + probability. The program can branch right away. The official claim: this kind of closed decision is dozens to two hundred times faster than large models, with input about
Show the original post in its source language ▾
刚刷到一个很野的新模型：**Jev**。 它不聊天、不写文、不写代码。 你丢给它一段状态 + 几个你自己定好的问题，它只干一件事——**做判断**。 选哪个队列、紧急几分、要不要人工、下一步点哪个按钮，直接返回选项 + 概率。程序拿来就能分支。 官方说法：这类封闭决策比大模型快几十到两百倍，输入约
The post above is a machine translation from zh; the untranslated text is in the fold-out.
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.
Where this case fits
Filed under general Jev builds. In the pattern Jev is built for, the model answers a bounded question per step — and ordinary code acts on the answer, because the answer is already a value rather than a paragraph. Browse the full case library for the work it sits next to.
Keep browsing: all 1173 Jev cases · more from @gaoqian2580 · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post