Kandinsky LabKandinsky control models
Kandinsky 3 ControlNet HED
Kandinsky's developers describe a trained HED-based ControlNet for image conditioning. Its weights are published through ai-forever.
Research release
Use the published package with its matching implementation
The weight repository has no descriptive model card. Architecture, training-set details and a checkpoint-specific license could not be confirmed from that repository.ai-forever/kandinsky3_controlnet_hed — model card and files
Treat the condition as a structural input
Practical guidance. The conditioning type identifies the intended guidance signal. It does not establish the model's quality on arbitrary drawings or depth maps.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | ai-forever/kandinsky3_controlnet_hed |
| Condition | HED edgeai-forever/kandinsky3_controlnet_hed — model card and files |
| Weight license | Not stated in the checked weight repository.ai-forever/kandinsky3_controlnet_hed — model card and files |
Related models
- Kandinsky 3.0
Kandinsky Lab · Kandinsky 3
An image diffusion model with a larger text encoder and U-Net, trained with attention to Russian cultural subjects.
Launch date not confirmedDownloadable