You're looking at a specific version of this model. Jump to the model overview.
codehappynice /infinitetalk:9f0d42a7
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| image |
string
|
Reference portrait image
|
|
| audio |
string
|
Driving audio (wav/mp3/m4a)
|
|
| prompt |
string
|
A person is talking naturally to the camera.
|
Describe the scene and motion
|
| negative_prompt |
string
|
bright tones, overexposed, static, blurred details, subtitles, style, works, paintings, images, static, overall gray, worst quality, low quality, JPEG compression residue, ugly, incomplete, extra fingers, poorly drawn hands, poorly drawn faces, deformed, disfigured, misshapen limbs, fused fingers, still picture, messy background, three legs, many people in the background, walking backwards
|
Negative prompt
|
| resolution |
None
|
480
|
Output resolution bucket
|
| num_frames |
integer
|
81
Min: 25 Max: 1001 |
Number of frames (auto-adjusted to 4n+1). 81 frames ~= 3.2s
|
| sampling_steps |
integer
|
40
Min: 4 Max: 100 |
Diffusion sampling steps
|
| text_guide_scale |
number
|
5
Min: 1 Max: 15 |
Text guidance scale (CFG). Optimal value is ~5 without LoRA.
|
| audio_guide_scale |
number
|
4
Min: 1 Max: 15 |
Audio guidance scale (lip sync). Higher improves sync; ~4 is optimal.
|
| use_teacache |
boolean
|
True
|
Enable TeaCache acceleration
|
| seed |
integer
|
Random seed for reproducible generation. Leave blank to randomize.
|
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{'format': 'uri', 'title': 'Output', 'type': 'string'}