divamgupta / diffusionbee-stable-diffusion-ui

Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed.
https://diffusionbee.com
GNU Affero General Public License v3.0
12.51k stars 619 forks source link

add Kandinsky 2.1 - the first multilingual model #410

Open 0-NiK-0 opened 1 year ago

0-NiK-0 commented 1 year ago

Kandinsky 2.1 https://github.com/ai-forever/Kandinsky-2 https://huggingface.co/ai-forever/Kandinsky_2.0

Info https://habr.com/ru/companies/sberbank/articles/725282/ https://www.reddit.com/r/StableDiffusion/comments/12bf5k2/kandinsky_21_beats_stable_diffusion_and_allows/

Demo https://fusionbrain.ai/ https://t.me/kandinsky21_bot https://rudalle.ru/kandinsky2 https://huggingface.co/spaces/ai-forever/Kandinsky2.1

Model architecture: It is a latent diffusion model with two multilingual text encoders: mCLIP-XLMR 560M parameters mT5-encoder-small 146M parameters These encoders and multilingual training datasets unveil the real multilingual text-to-image generation experience!

227786225-55d3a31b-28a1-4a30-acc2-2bbae33bd188 kandinsky21
Balackburn commented 1 year ago

You can add custom model

twi123 commented 1 year ago

You can add custom model

Can you, pls, provide step-by-step guide for using Diffusion Bee with Kandinsky 2.1 model? https://huggingface.co/ai-forever/Kandinsky_2.1/tree/main - Kandinsky 2.1 .ckpt files https://huggingface.co/dreamlike-art/kandinsky-2.1/tree/main - Kandinsky 2.1 files converted to .safetensors

I've trying by myself with no luck...

Balackburn commented 1 year ago

@twi123 If this model is not based on stable distribution, it won't work.