ComfyUI Node
Audio Gen Video 🎥AniPortrait
A ComfyUI node in AniPortrait 🎥Video with 20 inputs and 1 output.
Audio Gen Video 🎥AniPortrait
- ref_image
- images
- audio_path
- images
â—„height512â–º
â—„width512â–º
â—„seed42â–º
â—„cfg3.5â–º
â—„steps25â–º
◄vae_path▾►
◄model▾►
◄weight_dtype▾►
â—„acceleratetrueâ–º
â—„length0â–º
â—„fi_step3â–º
◄motion_module_path▾►
◄image_encoder_path▾►
◄denoising_unet_path▾►
◄reference_unet_path▾►
◄pose_guider_path▾►
â—„fps0â–º
CategoryAniPortrait 🎥Video
Inputs (20)
| Name | Type | Default | Description |
|---|---|---|---|
| ref_image | IMAGE | — | |
| height | INT | 5120–1024 | — |
| width | INT | 5120–1024 | — |
| seed | INT | 42 | — |
| cfg | FLOAT | 3.50–10 | — |
| steps | INT | 250–50 | — |
| vae_path | COMBO | 1 options: pretrained_model/sd-vae-ft-mse | |
| model | COMBO | 1 options: pretrained_model/stable-diffusion-v1-5 | |
| weight_dtype | COMBO | 2 options: fp16, fp32 | |
| accelerate | BOOLEAN | true | — |
| length | INT | 00–18446744073709550000 | — |
| fi_step | INT | 3 | — |
| motion_module_path | COMBO | 1 options: pretrained_model/motion_module.pth | |
| image_encoder_path | COMBO | 1 options: pretrained_model/image_encoder | |
| denoising_unet_path | COMBO | 1 options: pretrained_model/denoising_unet.pth | |
| reference_unet_path | COMBO | 1 options: pretrained_model/reference_unet.pth | |
| pose_guider_path | COMBO | 1 options: pretrained_model/pose_guider.pth | |
| imagesopt | IMAGE | — | |
| audio_pathopt | Audio_Path | — | |
| fpsopt | INT | 00–18446744073709550000 | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| images | IMAGE | — |