Article summary: A configuration that uses an LLM in Ollama
Replicated Jev's judge-only approach locally with Ollama, measuring a median 55ms and fastest 23ms on an RTX 5060 Ti.
The original post, translated
Article summary: A configuration that uses an LLM in Ollama only for "judgment," not "generation." In an RTX 5060 Ti environment, 15 measurements recorded a median of 55ms and a fastest time of 23ms. The author's claim is to reproduce locally the same idea as Jev (TypeSafe).
Show the original post in its source language ▾
記事の概要：OllamaでLLMを「生成」ではなく「判断」だけに使う構成。 RTX 5060 Ti環境で15回計測した結果、中央値55ms・最速23msを記録したとのこと。 Jev（TypeSafe）と同じ発想をローカルで再現する、というのが著者の主張。
The post above is a machine translation from ja; 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 @mikaeru676523 · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post