fix readme

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,54 @@
+---
+license: apache-2.0
+tags:
+- Kandinsky
+- text-image
+- text2image
+- diffusion
+- latent diffusion
+- mCLIP-XLMR
+- mT5
+---
+
+# Kandinsky 2.1
+
+
+[Open In Colab](https://colab.research.google.com/drive/1xSbu-b-EwYd6GdaFPRVgvXBX_mciZ41e?usp=sharing)
+
+[GitHub repository](https://github.com/ai-forever/Kandinsky-2)
+
+[Habr post](https://habr.com/ru/company/sberbank/blog/725282/)
+
+[Demo](https://rudalle.ru/)
+
+## Architecture
+
+Kandinsky 2.1 inherits best practicies from Dall-E 2 and Latent diffucion, while introducing some new ideas.
+
+As text and image encoder it uses CLIP model and diffusion image prior (mapping) between latent spaces of CLIP modalities. This approach increases the visual performance of the model and unveils new horizons in blending images and text-guided image manipulation.
+
+For diffusion mapping of latent spaces we use transformer with num_layers=20, num_heads=32 and hidden_size=2048.
+
+![](https://raw.githubusercontent.com/ai-forever/Kandinsky-2/main/content/kandinsky21.png)
+
+Other architecture parts:
+
++ Text encoder (XLM-Roberta-Large-Vit-L-14) - 560M
++ Diffusion Image Prior — 1B
++ CLIP image encoder (ViT-L/14) - 427M
++ Latent Diffusion U-Net - 1.22B
++ MoVQ encoder/decoder - 67M
+
+
+![](https://raw.githubusercontent.com/ai-forever/Kandinsky-2/main/content/einstein.png)
+
+
+# Authors
+
++ Arseniy Shakhmatov: [Github](https://github.com/cene555), [Blog](https://t.me/gradientdip)
++ Anton Razzhigaev: [Github](https://github.com/razzant), [Blog](https://t.me/abstractDL)
++ Aleksandr Nikolich: [Github](https://github.com/AlexWortega), [Blog](https://t.me/lovedeathtransformers)
++ Vladimir Arkhipkin: [Github](https://github.com/oriBetelgeuse)
++ Igor Pavlov: [Github](https://github.com/boomb0om)
++ Andrey Kuznetsov: [Github](https://github.com/kuznetsoffandrey)
++ Denis Dimitrov: [Github](https://github.com/denndimitrov)

README.md

@@ -10,41 +10,38 @@ tags:
 - mT5
 ---
 
-# Kandinsky 2.0
+# Kandinsky 2.1
 
 
-Kandinsky 2.0 — the first multilingual text2image model.
+[Open In Colab](https://colab.research.google.com/drive/1xSbu-b-EwYd6GdaFPRVgvXBX_mciZ41e?usp=sharing)
 
-[Open In Colab](https://colab.research.google.com/drive/1uPg9KwGZ2hJBl9taGA_3kyKGw12Rh3ij?usp=sharing)
+[GitHub repository](https://github.com/ai-forever/Kandinsky-2)
 
-[GitHub repository](https://github.com/ai-forever/Kandinsky-2.0)
-
-[Habr post](https://habr.com/ru/company/sberbank/blog/701162/)
+[Habr post](https://habr.com/ru/company/sberbank/blog/725282/)
 
 [Demo](https://rudalle.ru/)
 
-**UNet size: 1.2B parameters**
+## Architecture
+
+Kandinsky 2.1 inherits best practicies from Dall-E 2 and Latent diffucion, while introducing some new ideas.
 
-![NatallE.png](https://s3.amazonaws.com/moonup/production/uploads/1669132577749-5f91b1208a61a359f44e1851.png)
+As text and image encoder it uses CLIP model and diffusion image prior (mapping) between latent spaces of CLIP modalities. This approach increases the visual performance of the model and unveils new horizons in blending images and text-guided image manipulation.
 
-It is a latent diffusion model with two multi-lingual text encoders:
-* mCLIP-XLMR (560M parameters)
-* mT5-encoder-small (146M parameters)
+For diffusion mapping of latent spaces we use transformer with num_layers=20, num_heads=32 and hidden_size=2048.
 
+![](https://raw.githubusercontent.com/ai-forever/Kandinsky-2/main/content/kandinsky21.png)
 
-These encoders and multilingual training datasets unveil the real multilingual text2image generation experience!
+Other architecture parts:
 
-![header.png](https://s3.amazonaws.com/moonup/production/uploads/1669132825912-5f91b1208a61a359f44e1851.png)
++ Text encoder (XLM-Roberta-Large-Vit-L-14) - 560M
++ Diffusion Image Prior — 1B
++ CLIP image encoder (ViT-L/14) - 427M
++ Latent Diffusion U-Net - 1.22B
++ MoVQ encoder/decoder - 67M
 
-# How to use
 
-```python
-pip install "git+https://github.com/ai-forever/Kandinsky-2.0.git"
+![](https://raw.githubusercontent.com/ai-forever/Kandinsky-2/main/content/einstein.png)
 
-from kandinsky2 import get_kandinsky2
-model = get_kandinsky2('cuda', task_type='text2img')
-images = model.generate_text2img('кошка в космосе', batch_size=4, h=512, w=512, num_steps=75, denoised_type='dynamic_threshold', dynamic_threshold_v=99.5, sampler='ddim_sampler', ddim_eta=0.01, guidance_scale=10)
-```
 
 # Authors
 
@@ -54,7 +51,4 @@ images = model.generate_text2img('кошка в космосе', batch_size=4, h
 + Vladimir Arkhipkin: [Github](https://github.com/oriBetelgeuse)
 + Igor Pavlov: [Github](https://github.com/boomb0om)
 + Andrey Kuznetsov: [Github](https://github.com/kuznetsoffandrey)
-+ Denis Dimitrov: [Github](https://github.com/denndimitrov)
-
-
-
++ Denis Dimitrov: [Github](https://github.com/denndimitrov)