Post

amd halo AI/strix halo 128GB inference box with customized Qwen3.8 Flash-Next Strata settings, built on NixOS+K3s. Basically ChatGPT in a flake. github.com/devon-system...

github.comGitHub - devon-systems/kanto: Strix Halo ML inference + infrastructure IaC for NixOS+K3sStrix Halo ML inference + infrastructure IaC for NixOS+K3s - devon-systems/kanto

aly

@aly.codes

Designed this way bc I intend to add another one to the cluster :3

25/10/26, 9:51 pm0 reposts7 likes

jae (hallowed)

@fubarchitect.com

hell yeah