ComfyUI Node
FunUI Identity Transfer
Carry a reference face into an LTX clip (Best-FaceID style): the reference goes in as separate tokens the model attends to, plus optional ArcFace text tokens. With no image wired, everything passes through untouched.
FunUI Identity Transfer
- model
- vae
- positive
- negative
- latent
- image
- model
- positive
- negative
- status
◄identity_projectorNone►
◄source_id2►
◄phase_scale1.0►
◄id_strength1.0►
◄arcface_modeauto_adjust►
CategoryFunUI/Conditioning
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| vae | VAE | The picture VAE, to encode the reference. | |
| positive | CONDITIONING | — | |
| negative | CONDITIONING | — | |
| latent | LATENT | The clip's starting latent: the reference is sized to its frame. | |
| imageopt | IMAGE | The reference face. None = no identity transfer. | |
| identity_projectoropt | COMBO | None | ArcFace projector .safetensors (models/loras). None = the overlap tokens only (they carry most of the identity). |
| source_idopt | FLOAT | 20–8 | Source-phase id for the reference tokens (the trainer used 2). 0 turns the rotation off, leaving the tokens in. |
| phase_scaleopt | FLOAT | 1.00–4 | — |
| id_strengthopt | FLOAT | 1.00–50 | Multiplies the ArcFace tokens. A weak channel: 5-20 to test. |
| arcface_modeopt | COMBO | auto_adjust | auto_adjust retries detection zoomed out/up; as_is detects on the image only; disable skips ArcFace. |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| model | MODEL | — |
| positive | CONDITIONING | — |
| negative | CONDITIONING | — |
| status | STRING | — |