ComfyUI Node
WanAnimatePlus ClipVisionEncode V2
A ComfyUI node in WanAnimatePlus with 9 inputs and 1 output.
WanAnimatePlus ClipVisionEncode V2
- clip_vision
- images
- negative_image
- image_embeds
◄strength1.000►
◄cropcenter►
◄combine_embedsaverage►
◄force_offloadtrue►
◄tiles0►
◄ratio0.50►
CategoryWanAnimatePlus
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| clip_vision | CLIP_VISION | — | |
| images | IMAGE | Image sequence to encode. All frames in the IMAGE batch are accepted. | |
| strength | FLOAT | 1.0000–10 | Additional multiplier applied to every image embed |
| crop | COMBO | center | Crop image to 224x224 before encoding |
| combine_embeds | COMBO | average | Method to combine multiple clip embeds |
| force_offload | BOOLEAN | true | — |
| negative_imageopt | IMAGE | Image batch to use for uncond | |
| tilesopt | INT | 00–16 | Use matteo's tiled image encoding for improved accuracy |
| ratioopt | FLOAT | 0.500–1 | Ratio of the tile average |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| image_embeds | WANVIDIMAGE_CLIPEMBEDS | — |