devwithjev
— reading now— views
Submit a build

Jev-Powered Chess via Computer Use

The author connected Jev to Codex and had Astra play chess using Computer Use, with Astra running independently as an A/B control.

View on X cost额度消耗多出2倍time用时反而更慢
洞寓法师🦁@IFITALEX𝕏
如果你也是这两天被 X 时间线上的 Jev 模型刷屏、正跃跃欲试的 AI 实践者:我刚替你踩了一个价值 33% 周额度的大坑😭 先说我的结论:盲目把jev模型塞进 Agent 流程,很可能会换来翻倍的账单和更拖沓的执行。 为了验证 Jev 模型的实战表现,我把它接入 Codex,希望配合 Astra 做 Computer Use 下国际象棋 ♟️。我让 Astra 独立执行作为 A/B 对照组,跑完这轮实测,我苦笑不得:额度消耗多出2倍,用时反而更慢。 再看许多Jev的成功用例,我才发现,Jev 真正的舒适区,不在于像国际象棋这种,需要深度逻辑的推理环节,而在于快速思考的执行层。
Sep 22, 2026X postsView on X
In the test, the Jev setup used more quota and took longer. The author concluded that Jev is better suited to fast execution than to reasoning tasks requiring deep logic, such as chess.

Also filed under Games & real time