rinnakk / japanese-stable-diffusion

Japanese Stable Diffusion is a Japanese specific latent text-to-image diffusion model capable of generating photo-realistic images given any text input.

rinnakk/japanese-stable-diffusion Issues

Do you have any plan to publish Japanese caption for training data?
Updated a year ago
Where can I get vocab.json of "japanese-stable-diffusion"'s tokenizer
Closed a year ago1
KeyError in scripts/txt2img.ipynb
Closed a year ago3
How did you setup the loss for the unet in training?
Updated 2 years ago
The hardware and training information
Updated 2 years ago
How to train the first stage?
Updated 2 years ago15
Fintuning code?
Updated 2 years ago