4 ms·

I'm running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I've had some success with LLMs, but only the smaller m

I'm running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I've had some success with LLMs, but only the smaller models. No idea where to start with images, everything seems geared towards msft+nvda.

collinvandyck76 2y ago
DiffusionBee will let you do this quite easily. edit: nevermind, it's a macos app
- lagniappe 2y ago
  Is DiffusionBee still in development? I had stopped using it because it seemed like the dev interest had stalled.
  - collinvandyck76 2y ago
    It gets periodic releases, but the source isn't typically updated at the same time.
LeoPanthera 2y ago
"Draw Things" is a native Mac app for text to image. It's a a lot more advanced than DiffusionBee, it will download the models for you, and it's free. It's also available for iOS. (!)
- smcleod 2y ago
  Draw things is neat but it's so damn slow compared to other tools (e.g. invokeai), I'm not sure why it takes so long to generate images with any model?
  - LeoPanthera 2y ago
    It's not any slower than invokeai for me. Maybe check the settings, and try using the GPU instead of CoreML.
  - liuliu 2y ago
    On the same Mac hardware, Draw Things should be the fastest on models such as SDXL / FLUX.1 against other tools based on PyTorch (I stopped benchmarking SD v1.5 results for a while so that might regress a little bit here or there).
loxias 2y ago
Try https://github.com/leejet/stable-diffusion.cpp https://github.com/leejet/stable-diffusion.cpp