Kandinsky-2

Kandinsky-2 copied to clipboard

Mixed CPU/GPU generation on 2.2

1

Hello! This is not actually an issue, but a kind of a "how-to" post, namely how to use the model on a lower-end GPUs. The way proposed here is to...

trolley813

Fine tuning guide

2

A guide of how to fine tune the model to improve it

MrlolDev

Huggingface 2.2 model?

1

where is the link for kandinsky 2.2 at huggingfacE?

MrlolDev

controlnet

1

liangwq

Могу я использовать Kandinsky с видеокартой gtx 1650? Попробовал установить: пыхтит, гудит, но при разрешении больше примерно 100х100 получаются только чёрные квадраты Параметры запуска мб там какие-то есть?

nahuyaqa

transformers being locked to 4.23.1 causing issues

1

In setup.py, is there a reason the transformers dependency is locked to 4.23.1? I'm hitting an error trying to build a docker worker image of Kandinsky2 with img2img support. There...

xelfer

Added callbacks, Mac Compatibility added Added model offloads to cpu (tested by @Thaevil1 on RTX A2000- 512x512: 6Gb VRAM) You can test my fork of Kandinsky utilizing these changes in...

XmYx

CUDA support

4

How can run it on machine without Nvidia Graphics Card?

imdigitalashish

This is really a surprising work. I am very curious about the composition of training data.I would greatly appreciate it if you could share the full details of the training...

Sun-Shiqi

Hi, I have a question about using the DDIM inversion scheduler with Kandinsky 2_1. Can it be directly operated with [this method](https://github.com/ai-forever/Kandinsky-2/blob/47c01fa310ce036da46b6335d9f99e66edfdaab2/kandinsky2/model/gaussian_diffusion.py#L521)? ```python ... @torch.no_grad() def test( self, num_steps =...

tikitong