Repository and full retrospective: https://
Built Necro, a local decision model with a TypeSafe SDK-compatible interface, and benchmarked it against Jev on context, question, and candidate answers.
The original post, translated
Repository and full retrospective: https:// github.com/FogMoe/necro Necro wants to build a local judgment model similar to Jev, with an interface compatible with the TypeSafe SDK, and also used Jev for evaluation comparison. It takes input context, question, and candidate answers, and directly outputs the choice and probability. I want to use this project to explain GPT-6 Astra
Show the original post in its source language ▾
仓库和完整复盘： https:// github.com/FogMoe/necro Necro 想做一个类似 Jev 的本地判断模型，接口兼容 TypeSafe SDK，也用 Jev 做了评测对照。输入上下文、问题和候选答案，直接输出选择及概率。 我想通过这个项目说明 GPT-6 Astra
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 coding & developer tools. 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 coding & developer tools page.
Related Jev cases
I let GPT-6 Astra take full charge of a Jev-like model
Dev tools
I'm distilling jev, wait for my good news.
The gains aren’t free: Jev can't generate text Comparing
Dev toolsClassification
Jev VS Fable 5.1 VS GPT-6 Astra: chess typesafe's new Jev
GamesDev tools
Keep browsing: all 1173 Jev cases · more from @ScarletKc · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post