jevVIDEO
Laya-MLX:Apple 芯片上的开源 Jev 式决策模型,Snake demo 每秒 11 次决策
Laya 的 MLX 移植版在 Mac Studio M4 Max 上跑 Snake demo,每秒约 11 次决策,全部本地推理,零输出 token。
Kai
@hqmank2026.09.22收录原推发布于 2026.09.21
Kai @hqmank原推发布于 2026.09.21
Following Jev? Laya-MLX is worth a look. It's an open-source MLX port of Laya for Apple silicon. You give it text and possible actions; it returns probabilities for each, with zero output tokens. I ran its Snake demo on my Mac Studio M4 Max and recorded the clip below. Around 11 decisions/sec, all inference on my Mac. The demo calls the model for each move, with a planner and safety layer assisting. Credit to @mizorewww for the MLX port. Repo in reply 👇
x.com/hqmank/status/2102035210506207258采集于 2026.09.22 10:21