tgohblio/instant-id-multicontrolnet

Use pose for skeleton inference

Seed number. Set to non-zero to make the image reproducible.

Use canny for edge detection

Select SDXL model

Input prompt

Use depth for depth map estimation

Number of denoising steps. If enable fast mode, this is not used.

Scheduler options. If enable fast mode, this is not used.

Scale for classifier-free guidance. Optimum is 4-8. If enable fast mode, this is not used.

Safety checker is enabled by default. Un-tick to expose unfiltered results.

Image of your face

if enable fast mode, choose number of denoising steps

Input negative prompt

Reference pose image

Enable SDXL-lightning fast inference. If pose, canny or depth map is used, disable it for better quality images.

Image adapter strength (for detail)

Enhance non-face region

IdentityNet strength (for fidelity)