@CleverMoniker done with anything-v4.5-pruned-fp32.ckpt
took me 15 mins to render, cpu only (AMD Ryzen 9 3900X)
prompts:
masterpiece, best quality, 1girl, fox ears, (fox tail:1.2), blonde hair, red eyes, freckles, arms behind head, arms up, on bed, red dress
Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, furry, multiple tails, nude
Steps: 30, Sampler: Euler a, CFG scale: 9, Seed: 1697743345, Size: 720x960, Model hash: 1d10a19f06, Model: anything-v4.5-pruned-fp32
cc @skylar @theorytoe @Giganova8
took me 15 mins to render, cpu only (AMD Ryzen 9 3900X)
prompts:
masterpiece, best quality, 1girl, fox ears, (fox tail:1.2), blonde hair, red eyes, freckles, arms behind head, arms up, on bed, red dress
Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, furry, multiple tails, nude
Steps: 30, Sampler: Euler a, CFG scale: 9, Seed: 1697743345, Size: 720x960, Model hash: 1d10a19f06, Model: anything-v4.5-pruned-fp32
cc @skylar @theorytoe @Giganova8
Follow
@bronze @theorytoe @CleverMoniker @Giganova8 @skylar how did you make this !?
@bronze @VooDooMedic @theorytoe @CleverMoniker @Giganova8 @skylar I got stable diffusion working for me. but it takes forever since I had to disable the GPU processing since apparently my 4 year old gpu is to old and im running windows. do you have any experience with AMD gpu on windows
@Groomschild @theorytoe @VooDooMedic @CleverMoniker @Giganova8 @skylar
>windows
you lost me there lol
Apparently, theres support for AMD cards that can use ROCm, but that docker container I linked only does CPU only or nvidia CUDA.
I might be able to use my RX480 on my gaming desktop, but im not putting fucking docker on it and bloating up my nice little system and I'm also not installing 50000 python dependencies in a virtualenv or whatever. My desktop is messy as it is...
I'll just render shit overnight on my ryzen boxes. Easier to justify docker on something headless anyways.
Your best luck would be to start dual booting linux already. Just install whichever distro catches your eye and roll with it for a bit, whatever you use will work 1000x better than anything on windows.
>windows
you lost me there lol
Apparently, theres support for AMD cards that can use ROCm, but that docker container I linked only does CPU only or nvidia CUDA.
I might be able to use my RX480 on my gaming desktop, but im not putting fucking docker on it and bloating up my nice little system and I'm also not installing 50000 python dependencies in a virtualenv or whatever. My desktop is messy as it is...
I'll just render shit overnight on my ryzen boxes. Easier to justify docker on something headless anyways.
Your best luck would be to start dual booting linux already. Just install whichever distro catches your eye and roll with it for a bit, whatever you use will work 1000x better than anything on windows.
@bronze @theorytoe @VooDooMedic @CleverMoniker @Groomschild @Giganova8 i run it on my desktop, but i already had the snake app
@bronze @theorytoe @VooDooMedic @CleverMoniker @Groomschild @Giganova8 python
(docker is the whale app)
(docker is the whale app)
@skylar @theorytoe @VooDooMedic @CleverMoniker @Groomschild @Giganova8
snake app is the bane of my existence
t. gentoo user
snake app is the bane of my existence
t. gentoo user
got a spare computer / server? use linux and docker.
https://github.com/AbdBarho/stable-diffusion-webui-docker
https://github.com/AbdBarho/stable-diffusion-webui-docker/wiki/Setup instructions
you can find models here:
https://huggingface.co/models?other=stable-diffusion
keep in mind these are huge files, youre gonna need a lot of hard drive space
you can download them directly or use git (with the lfs addon) to download the whole model's git repo (full of variants of the same model)
put the file in ./data/StableDiffusion/ or use docker to symlink it to that directory inside the container like so
(in the docker-compose.yml file, under either the auto or auto-cpu service)
volumes:
- /home/user/models/anything-v4.0/anything-v4.5-pruned-fp32.ckpt:/data/StableDiffusion/anything-v4.5-pruned-fp32.ckpt
that way you dont have to move or copy the model over to whichever folder your docker container is running from
HAVE FUN!