ComfyUI Node
VAE Decode Tiled (Batched)
Same as the built-in VAE Decode (Tiled), but adds a batch_size field that controls how many latents are decoded by the VAE at once. Decoding fewer at a time (e.g. 1) can lower peak VRAM.
VAE Decode Tiled (Batched)
- samples
- vae
- IMAGE
◄tile_size512►
◄overlap64►
◄temporal_size64►
◄temporal_overlap8►
◄batch_size1►
CategoryvsLinx/latent
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| samples | LATENT | — | |
| vae | VAE | — | |
| tile_size | INT | 51264–4096 | — |
| overlap | INT | 640–4096 | — |
| temporal_size | INT | 648–4096 | Only used for video VAEs: Amount of frames to decode at a time. |
| temporal_overlap | INT | 84–4096 | Only used for video VAEs: Amount of frames to overlap. |
| batch_size | INT | 11–4096 | How many latents to hand to the VAE per decode call. Lower values reduce peak VRAM at the cost of more sequential calls; the default of 1 decodes one image at a time. Values >= the batch size behave exactly like the built-in node (single decode). |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| IMAGE | IMAGE | — |