Type: huggingface.image_to_image.VAEEncode
Namespace: huggingface.image_to_image
Description
Encodes an image into latents using a VAE. image -> tensor (TorchTensor)
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| model | hf.vae |
The VAE model to use. | {'type': 'hf.vae', 'repo_id': '', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None} |
| image | image |
Input image to encode. | {'type': 'image', 'uri': '', 'asset_id': None, 'data': None} |
| scale_factor | float |
Scaling factor applied to latents (e.g., 0.18215 for SD15) | 0.18215 |
Outputs
| Output | Type | Description |
|---|---|---|
| output | torch_tensor |
Metadata
Related Nodes
Browse other nodes in the huggingface.image_to_image namespace.