ComfyUI Node
Load Image Path (FFMPEGA)
Zero-memory image loader — outputs a file path STRING instead of an IMAGE tensor. Connect to FFMPEGA Agent's image_path slots. Uses ~0 MB for any number of images vs ~6 MB per image with standard Load Image. Accepts optional upstream IMAGE and image_path inputs for chaining.
Load Image Path (FFMPEGA)
- images
- image_path
- mask_points
- images
◄image▾►
◄image_path—►
◄mask_points—►
CategoryFFMPEGA
Inputs (4)
| Name | Type | Default | Description |
|---|---|---|---|
| image | COMBO | Select an image from ComfyUI's input directory or upload a new one. | |
| imagesopt | IMAGE | Optional upstream IMAGE input (e.g. from another node). Passed through as output. | |
| image_pathopt | STRING | Optional upstream image path. Overrides the file picker when connected. | |
| mask_pointsopt | STRING | Optional upstream mask_points pass-through. When connected, overrides the locally drawn mask points. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| image_path | STRING | Absolute file path to the selected image. Connect to FFMPEGA Agent's image_path_a / image_path_b / … inputs. |
| mask_points | STRING | JSON-encoded point selection data from the Point Selector. Connect to FFMPEGA Agent's mask_points input for guided masking. |
| images | IMAGE | Upstream IMAGE pass-through (or empty tensor if not connected). |