Does GPU Cloud service support Illyasviel/Fooocus AI?
My pc has low vram and always get disconnections from the fooocus ai, im interested to upgrade with a runpod gpu service, does it support the https://github.com/lllyasviel/Fooocus service?
GitHub
GitHub - lllyasviel/Fooocus: Focus on prompting and generating
Focus on prompting and generating. Contribute to lllyasviel/Fooocus development by creating an account on GitHub.
29 Replies
GitHub
GitHub - justinwlin/FooocusRunpod
Contribute to justinwlin/FooocusRunpod development by creating an account on GitHub.
Also many others under #🧩｜templates That ik @Brainpod Made some cool ones with presets and stuff already!
Mine is pretty barebones - it works as a clean one
I saw brainpod has many + this seems to be a newer one from them:
https://discord.com/channels/912829806415085598/1141049171726573620/1141049171726573620
I do think mine is as close to pure as u can get, which is nice, b/c less things to go wrong. Lol, but others definetly have benefits such as having anime style, realistic style, all that already ready.
so it's basically plug and play after the subscription through browser link? im kinda new to such stuff. Also, whats the recommended gpu/vram to use from the listing, am i covered with 24GB?
Can check out the README in my documentation
But I find that it works even at the cheapest GPU
The fooocus model doesn't seem to eat too much... maybe depends on the model ur using etc.
its higher quality and it usually crashes on the official collab page so yes
Runpod Link: https://runpod.io/gsc?template=8q4vl4gfld&ref=wqryvm1m
Hmmmm. not sure what u mean
But I think give it a try
I can usually run it on the lowest one
I usually run it on this on secure cloud
i mean on the google colab page outside the runpod using my original gpu which operates at max 4gb vram
If u have an issue can keep bumping it up
I see
ill message soon to check these stuff out first thx for the info
does using stable diffusion software require a cloud storage or i run everything from my pc using the external gpu? Also when im not using the GPU is there a way to pause the charging of the service while i dont use it?
There is a stop button to pause it
if u have a network storage, it only lets u terminate it
since u can store everything under /workspace
and will persis to the network volume
Ur essentially running it from the cloud
u aren't using ur local pc
ur using a remote server
with a gpu
and u gotta get ur files to that server
What about the other specs?
wym?
like cpu / ram?
aren't controllable
Do I use my own RAM CPU?
no
its all a separate computer in the sky
that ur connecting to
nothing is local
Oh nice and I guess I have quality speed as well?
Fast generation for example
SHould be decent, the pods tell u how fast the download / upload is
depends on the gpu ur using but yes
that is the goal
depends on the setup too
for ex.
SECourses
YouTube
Double Your Stable Diffusion Inference Speed with RTX Acceleration ...
Stable Diffusion Gets A Major Boost With RTX Acceleration. One of the most common ways to use Stable Diffusion, the popular Generative AI tool that allows users to produce images from simple text descriptions, is through the Stable Diffusion Web UI by Automatic1111. In today’s Game Ready Driver, NVIDIA added TensorRT acceleration for Stable Diff...
@YouTube.com/@SECourses has great videos
on runpod
+ also increasing speeds
for different generation
there are other resources scattered around the discord
but yea
That's quite resourceful I'll check it out first
it just essentially if ur setup is good, and some optimizations, u can get different levels of speed
ill say it runs better than anything ill ever run on my mac that just dies xD
I generate a lot with Fooocus too, so i think is decent
Well makes sense if it's a Mac
yeh
So you're actually doing the work without much of a problem on the cloud as well
I'll research it more tomorrow cuz it's late now, but if it's all on the cloud then it works even better
Will be back soon
thanks for tagging
If i choose to deploy the fooocus template can i switch after with stable diffusion on the same cloud/deployment/drive or i need a new gpu/cloud to use another software/platform?
If u download stable diffusion to the same template it is possible
U can do this manually or make ur own docker
otherwise u beed to do it with a new gpu