zlacker

[parent] [thread] 4 comments
1. idontw+(OP)[view] [source] 2025-12-06 18:43:54
Does it run on apple silicon?
replies(2): >>iamfli+E1 >>sheeps+I1
2. iamfli+E1[view] [source] 2025-12-06 18:55:26
>>idontw+(OP)
It's working for me - it does max out my 64GB though.
replies(1): >>sheeps+K2
3. sheeps+I1[view] [source] 2025-12-06 18:55:42
>>idontw+(OP)
Apparently - https://github.com/ivanfioravanti/z-image-mps

Supports MPS (Metal Performance Shaders). Using something that skips Python entirely along with a mlx or gguf converted model file (if one exists) will likely be even faster.

replies(1): >>opensa+Hu
◧◩
4. sheeps+K2[view] [source] [discussion] 2025-12-06 19:04:21
>>iamfli+E1
Wow. I always forget how unlike autoregressive models, diffusion models are heavier on resources (for the same number of parameters).
◧◩
5. opensa+Hu[view] [source] [discussion] 2025-12-06 23:13:46
>>sheeps+I1
(Not tested) though apparently it already exists: https://github.com/leejet/stable-diffusion.cpp/wiki/How-to-U...
[go to top]