Jev just got popular a couple of days ago, and someone has
Trained Bespoke Nimble, an open Qwen3.5-9B model, in about two days using Jev only as an evaluation reference, not for distillation.
The original post, translated
Badass! Jev just got popular a couple of days ago, and someone has already made an open-source Jev🤣 @bespokelabsai released an open-source version of Jev: Bespoke Nimble. Even more absurd is that the entire project is based on Qwen3.5-9B and took only about two days. The data, model weights, training recipe, and inference code are all public, and it didn't distill from Jev, only used Jev as an evaluation reference. The final method is on Qwen3.5-9B
Show the original post in its source language ▾
牛逼！Jev 刚火没两天，已经有人搞了个开源的Jev🤣 @bespokelabsai 发布了一个开源版的Jev：Bespoke Nimble。 更离谱的是，整个项目是基于 Qwen3.5-9B ，只花了大约两天。 数据、模型权重、训练配方、推理代码全部公开，而且没有从 Jev 蒸馏，只拿 Jev 做评测参照。 最终方法是在 Qwen3.5-9B
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 @MaxForAI · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post