README
Kubin
Disclaimer
After 2.5 years, Kandinsky models are finally entering the mainstream generative AI ecosystem and starting to get the recognition they deserve. At this point, this project, its GUI and its overall approach to building ML tools, feels outdated and no longer reflects how this kind of software is typically built today. I may still push occasional updates, but no major development is planned here. The Kandinsky models themselves will for sure continue to appear in some of my other upcoming projects.
Description
Web-GUI for Kandinsky generative AI models (Kandinsky2, Kandinsky3, Kandinsky4, Kandinsky5).
Features
| Model | 2.0 | 2.1 | 2.2 | 3.0 | 3.1 | 4.0 | 5.0 |
|---|---|---|---|---|---|---|---|
| Txt2img | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| Img2img | ✔ | ✔ | ✔ | ✔ | 🕒 | ||
| Txt2Video | ✔ | ✔ | |||||
| Img2Video | 🕒 | ||||||
| Video2Audio | ✔ | ||||||
| Mixing | ✔ | ✔ | 🕒 | 🕒 | |||
| Inpainting | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| Outpainting | ✔ | ✔ | 🕒 | 🕒 | |||
| ControlNet | ✔ | 🕒 | |||||
| IpAdapter | 🕒 | ||||||
| Diffusers pipeline | ✔ | ✔ | ✔ | 🕒 | |||
| Low-VRAM options | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| ext. Image browser | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| ext. Image editor | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Fine-tuning | ✔ | 🕒 | |||||
| ext. LoRA training and inference | ✔ | 🕒 | |||||
| ext. Style library | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Image upscaling | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Video upscaling | 🕒 | 🕒 | |||||
| ext. Img2mesh | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Image interrogation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Video interrogation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Segmentation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Deforum | 🕒 | ✔ | |||||
| ext. Animation | ✔ | ✔ | |||||
| ext. Background remover | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| ext. LLM prompt enhancer | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| Dynamic prompts | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| Prompt weighting | 🕒 | ||||||
| ext. Area composition | 🕒 |
Screenshots
Toggle
Kandinsky 2.0
Kandinsky 2.1
Kandinsky 2.2
Kandinsky 3.0
Kandinsky 3.1
Kandinsky 4.0
Kandinsky 5.0
Google Colab
Documentation
Some bits of information can be found here.
Contributors
531 commits
9 commits
Similar repos
Languages
Python
96.9%
JavaScript
1.7%
README
Kubin
Disclaimer
After 2.5 years, Kandinsky models are finally entering the mainstream generative AI ecosystem and starting to get the recognition they deserve. At this point, this project, its GUI and its overall approach to building ML tools, feels outdated and no longer reflects how this kind of software is typically built today. I may still push occasional updates, but no major development is planned here. The Kandinsky models themselves will for sure continue to appear in some of my other upcoming projects.
Description
Web-GUI for Kandinsky generative AI models (Kandinsky2, Kandinsky3, Kandinsky4, Kandinsky5).
Features
| Model | 2.0 | 2.1 | 2.2 | 3.0 | 3.1 | 4.0 | 5.0 |
|---|---|---|---|---|---|---|---|
| Txt2img | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| Img2img | ✔ | ✔ | ✔ | ✔ | 🕒 | ||
| Txt2Video | ✔ | ✔ | |||||
| Img2Video | 🕒 | ||||||
| Video2Audio | ✔ | ||||||
| Mixing | ✔ | ✔ | 🕒 | 🕒 | |||
| Inpainting | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| Outpainting | ✔ | ✔ | 🕒 | 🕒 | |||
| ControlNet | ✔ | 🕒 | |||||
| IpAdapter | 🕒 | ||||||
| Diffusers pipeline | ✔ | ✔ | ✔ | 🕒 | |||
| Low-VRAM options | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| ext. Image browser | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| ext. Image editor | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Fine-tuning | ✔ | 🕒 | |||||
| ext. LoRA training and inference | ✔ | 🕒 | |||||
| ext. Style library | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Image upscaling | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Video upscaling | 🕒 | 🕒 | |||||
| ext. Img2mesh | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Image interrogation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Video interrogation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Segmentation | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| ext. Deforum | 🕒 | ✔ | |||||
| ext. Animation | ✔ | ✔ | |||||
| ext. Background remover | ✔ | ✔ | ✔ | ✔ | ✔ | ||
| ext. LLM prompt enhancer | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ |
| Dynamic prompts | ✔ | ✔ | ✔ | ✔ | ✔ | ✔ | |
| Prompt weighting | 🕒 | ||||||
| ext. Area composition | 🕒 |
Screenshots
Toggle
Kandinsky 2.0
Kandinsky 2.1
Kandinsky 2.2
Kandinsky 3.0
Kandinsky 3.1
Kandinsky 4.0
Kandinsky 5.0
Google Colab
Documentation
Some bits of information can be found here.
Similar repos
Contributors
531 commits
9 commits
Languages
Python
96.9%
JavaScript
1.7%