Kandinsky-2
Kandinsky-2 copied to clipboard
Kandinsky 2 — multilingual text2image latent diffusion model
Adding clip and opencv to the requirements
Some changes required to run it on CPU and other devices.
I installed Kandinsky according to the instructions: ``` PS D:\> pip install "git+https://github.com/ai-forever/Kandinsky-2.0.git" Collecting git+https://github.com/ai-forever/Kandinsky-2.0.git Cloning https://github.com/ai-forever/Kandinsky-2.0.git to c:\users\erel segal-halevi\appdata\local\temp\pip-req-build-o452dzak Running command git clone --filter=blob:none --quiet https://github.com/ai-forever/Kandinsky-2.0.git 'C:\Users\Erel Segal-Halevi\AppData\Local\Temp\pip-req-build-o452dzak' Resolved...
hi,dear just the first line code `>>> from kandinsky2 import get_kandinsky2` then got the error
Я не знаю куда правильно писать, т.к. Сбер не оставил вообще никаких контактов для обратной связи, поэтому пишу в единственное доступное место: сюда. Пользуюсь одноимённым ботом в ТГ и меня...
Infinite corridor effect appearing in two mirrors facing each other
How do I have to modify the settings so that it runs on my RTX3060ti card, because now I get OOM? It is a 8GB Nvidia Card, capable to run...
Is there a way for it to use the latent consistency model? Is it possible to run it on the CPU or use Kandinsky 3.0 as it is very fast...