ComfyUI Node
Fat Mex Face Swap
Face swap with two modes: • Quick – InsightFace inswapper paste (fast, no extra model needed) • Quality – YOLOv8 detect → inpaint face with editing model + 4x upscale Connect MODEL/CLIP/VAE from a Qwen Image Edit loader for quality mode.
Fat Mex Face Swap
- target_image
- face_reference
- model
- clip
- vae
- swapped_image
◄face_index0►
◄blend_strength1.00►
◄methodauto►
◄denoise0.60►
◄feather13►
◄steps4►
◄upscale_facetrue►
CategoryFat Mex
Inputs (12)
| Name | Type | Default | Description |
|---|---|---|---|
| target_image | IMAGE | The generated image to swap a face onto. | |
| face_reference | IMAGE | Clear photo of the face you want to use. | |
| modelopt | MODEL | For quality mode: connect MODEL from a Qwen Image Edit loader. | |
| clipopt | CLIP | For quality mode: connect CLIP from the same loader. | |
| vaeopt | VAE | For quality mode: connect VAE from the same loader. | |
| face_indexopt | INT | 00–10 | Which face to swap (0 = largest). |
| blend_strengthopt | FLOAT | 1.000–1 | Quick mode: blend strength. Quality mode: ignored. |
| methodopt | COMBO | auto | auto = quality if MODEL connected, else quick. |
| denoiseopt | FLOAT | 0.600.1–1 | Quality mode: how much to regenerate the face (0.6 = natural). |
| featheropt | INT | 130–50 | Quality mode: mask edge feathering in pixels. |
| stepsopt | INT | 41–30 | Quality mode: sampling steps for face inpaint. |
| upscale_faceopt | BOOLEAN | true | Quality mode: upscale face region with 4xFaceUpDAT. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| swapped_image | IMAGE | — |