Kandinsky LabKandinsky control models
Kandinsky 2.2 ControlNet Depth
This release adds depth conditioning to Kandinsky 2.2. The documented pipelines support text-to-image and image-to-image generation with the depth condition.
Provide the desired depth structure
Practical guidance. A depth condition can preserve broad geometry while the text specifies appearance. It is different from a reference-image identity condition or a mask that selects a replacement region.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | kandinsky-community/kandinsky-2-2-controlnet-depth |
| Weight license | Apache 2.0kandinsky-community/kandinsky-2-2-controlnet-depth — model card and files |
Related models
- Kandinsky 2.2
Kandinsky Lab · Kandinsky 2
A 1024-pixel image model with a retrained CLIP-based prior and an adapted diffusion decoder.
Launch date not confirmedDownloadable