ComfyUI Node
VAE Encode
A ComfyUI node in nifty/latent with 3 inputs and 2 outputs.
VAE Encode
- pixels
- vae
- LATENT
- IMAGE
◄target_latents0►
Categorynifty/latent
Inputs (3)
| Name | Type | Default | Description |
|---|---|---|---|
| pixels | IMAGE | Image frames to encode into latent space. | |
| vae | VAE | VAE model to use for encoding. | |
| target_latents | INT | 0-4096–4096 | 0 = encode all frames. Positive = encode only the first N latents worth of frames. Negative = encode only the last N latents worth of frames. The temporal compression factor is detected automatically. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| LATENT | LATENT | — |
| IMAGE | IMAGE | — |