Kandinsky LabKandinsky 5.0 Image
Kandinsky 5.0 Image Editing
This 6B image-editing checkpoint accepts an input picture and an instruction. It is the released editing model, separate from text-only image generation.
Apply an instruction to an existing picture
The editing line is published separately from the T2I line. Use its editing pipeline and source-image input. An image-generation checkpoint alone does not select the editing weights.kandinskylab/Kandinsky-5.0-I2I-Lite — model card and files
Load the corresponding image pipeline
The Diffusers export is identified as supervised fine-tuned. It uses Qwen2.5-VL and CLIP text encoders and a FLUX VAE. The native model table uses the matching SFT configuration.Kandinsky 5.0 I2I Lite SFT — Diffusers model cardkandinskylab/Kandinsky-5.0-I2I-Lite — model card and files
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | kandinskylab/Kandinsky-5.0-I2I-Litekandinskylab/Kandinsky-5.0-I2I-Lite-sft-Diffusers |
| Generator size | 6B parameterskandinskylab/Kandinsky-5.0-I2I-Lite — model card and files |
| Weight license | MITkandinskylab/Kandinsky-5.0-I2I-Lite — model card and files |
| Example image sizes | 1024 × 1024 and 1280 × 768kandinskylab/Kandinsky-5.0-I2I-Lite — model card and files |