ComfyUI Node
Load DiffDIS model by name
Auto download DiffDIS model to models/rembg
Load DiffDIS model by name
- vae
- positive
- model
- train_w
- train_h
- norm_params
◄model_name▾►
◄device▾►
◄dtypeAUTO►
CategoryRemBG_SET/Load
Inputs (5)
| Name | Type | Default | Description |
|---|---|---|---|
| model_name | COMBO | The name of the model, from the list of known models of this type | |
| device | COMBO | Device where the model will be run | |
| dtypeopt | COMBO | AUTO | Data type used for inference. `AUTO` means the same as the model on disk.Using a smaller size will save resources, but might degradate the results. |
| vaeopt | VAE | SD Turbo VAE for DiffDIS | |
| positiveopt | CONDITIONING | Experimental for DiffDIS |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| model | SET_REMBG | The remove background model ready to be used in a processing node |
| train_w | INT | Width of the images used to train this model |
| train_h | INT | Height of the images used to train this model |
| norm_params | NORM_PARAMS | Normalization parameters for the input images. This is needed only for advanced use when you want to manually pre-process the images. The `Arbitrary Normalize` node from `Image Misc` can use these parameters to apply the correct normalization. |