Running only the UI part locally while using external inference

[gpt4thewin]

Is this even a possibility ?
How dependent is AUTOMATIC1111 on nvidia ?

I would also like to run it on :
Environment: Rasberry pi (ARM platform, no Nvidia GPU) with Docker

• At the app level ?
• About running docker without the nvidia platform

[missionfloyd]

You can run it elsewhere and access it remotely with a raspberry pi.
https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Command-Line-Arguments-and-Settings#running-online

[gpt4thewin]

This looks like a possibility but then I need to run the expansive cloud setup when I use it.
I was thinking about using some pay-as-you-go inference

[KevinRussell]

I'd recommend looking into putting it on your own Jetson orin Nano? https://forums.developer.nvidia.com/c/agx-autonomous-machines/jetson-embedded-systems/jetson-orin-nano/632

[gpt4thewin]

Cool but it's an hardware investment