stable-diffusion
stable-diffusion copied to clipboard
A latent text-to-image diffusion model
PyTorch is defaulting to NVIDIA GPU, but it would be good to fall back to CPU-only if no suitable GPU is found. Something like this perhaps? https://github.com/ModeratePrawn/stable-diffusion-cpu/pull/1 This would provide...
It seems that the sd21-unclip-h.ckpt is damaged
Hi, I have a very big dataset that I want to test on stable diffusion. I want the model to be specialised in recognising the styles in my dataset. Just...
Hi, I have a use case where I believe fine-tuning the model with few of the params are freezed will be beneficial. I've modified the `init_from_ckpt` function in `ldm/models/diffusion/ddpm.py` as...
Hello, I’m looking to use the Stable Diffusion API to be able to remove objects from an image via a mask and have the inpainting tool magically fill in the...
Hi guys, I am running SD 1.5 locally on a machine with a 3060, with 12 Gb. txt2img is working flawlessly, but with img2img I often get 'corrupted' results. Below...
Hi, on the first install everything went pretty smoothly, the web user ui didn't give any problems until a certain one said she had a problem installing a package I...
## BUG of Proper Timestep Alignment in Sampling Methods such as DDIM Upon careful observation, I have discovered a significant issue that may appear trivial but is, in fact, anything...
Hi, I'm new to all of this so please bear with me. When I run nvidia-smi in conda to see if cuda is installed, I get back: `NVIDIA-SMI 426.78 Driver...
I had a fascinating question, why is it that if you type "a photograph of an astronaut riding a horse" or "a caat" you get the same result,