Jev hands-on test: browser decisions about 10 to 30 times

@smardio 麥客不停 · Sep 20, 2026 · case #200806 · listed via Jev Hub

Still from the clip posted with this case; watch it in the original post on X.

Benchmarked Jev against Hermes on three tasks, finding browser decisions 10-30x faster with 4-16x fewer output tokens.

The original post, translated

Jev hands-on test: browser decisions about 10 to 30 times faster, output tokens about 4 to 16 times fewer. Jev is a recently very popular model, what exactly is so magical about it? I had Hermes test it with three real tasks. Jev and common large models are not the same kind of thing. It is not responsible for writing articles, nor does it generate a passage of explanation; you give it the current state and question, and it only returns the answer and probability. It is more like giving an Agent

Show the original post in its source language ▾

Jev 实测：浏览器决策快约 10 到 30 倍，输出 token 少约 4 到 16 倍 Jev 是最近很火的模型，到底有什么神奇？我让 Hermes 用三个真实任务测了一遍。 Jev 和常见的大模型不是一类东西。它不负责写文章，也不生成一段解释；你把当前状态和问题交给它，它只返回答案和概率。它更像给 Agent

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

Engagement when collected

Likes0

Bookmarks0

Reposts0

Replies0

Quotes0

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 agents & workflows. 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. Other posts in the same family are on the agents & workflows page.

🤖 Agents & workflows