File size: 1,850 Bytes
e5b2e04 c86a694 684645c e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 0c7be1f e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 e5b2e04 1a6a960 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 |
---
license: apache-2.0
tags:
- Kandinsky
- text-image
inference: true
pipeline_tag: text-to-image
library_name: open_clip
---
# 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 diffusion, 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.
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
# 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) |