You're looking at a specific version of this model. Jump to the model overview.

tgohblio /instant-id-multicontrolnet:52039b70

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
face_image_path
string
Image of your face
pose_image_path
string
Reference pose image
prompt
string
a person
Input prompt
negative_prompt
string
(lowres, low quality, worst quality:1.2), (text:1.2), watermark, glitch, deformed, mutated, cross-eyed, ugly, disfigured, blurry, grainy
Input negative prompt
width
integer
640

Min: 512

Max: 2048

Width of output image
height
integer
640

Min: 512

Max: 2048

Height of output image
model
string (enum)
AlbedoBase XL V2

Options:

AlbedoBase XL V2, Juggernaut XL V8, Animagine XL V3, HelloWorld XL 5.0 GPT4V, SDXL RongHua V4

Select SDXL model
enable_fast_mode
boolean
True
Enable SDXL-lightning fast inference. If pose, canny or depth map is used, disable it for better quality images.
lightning_steps
string (enum)
4step

Options:

2step, 4step, 8step

if enable fast mode, choose number of denoising steps
scheduler
string (enum)
DPMSolverMultistepScheduler

Options:

DEISMultistepScheduler, HeunDiscreteScheduler, EulerDiscreteScheduler, DPMSolverMultistepScheduler, DPMSolverMultistepScheduler-Karras, DPMSolverMultistepScheduler-Karras-SDE

Scheduler options. If enable fast mode, this is not used.
adapter_strength_ratio
number
0.8

Max: 1

Image adapter strength (for detail)
identitynet_strength_ratio
number
0.8

Max: 1

IdentityNet strength (for fidelity)
pose
boolean
False
Use pose for skeleton inference
pose_strength
number
1

Max: 1.5

None
canny
boolean
False
Use canny for edge detection
canny_strength
number
0.5

Max: 1.5

None
depth_map
boolean
False
Use depth for depth map estimation
depth_strength
number
0.5

Max: 1.5

None
num_steps
integer
25

Min: 1

Max: 50

Number of denoising steps. If enable fast mode, this is not used.
guidance_scale
number
7

Max: 10

Scale for classifier-free guidance. Optimum is 4-8. If enable fast mode, this is not used.
seed
integer
0

Max: 2147483647

Seed number. Set to non-zero to make the image reproducible.
enhance_non_face_region
boolean
True
Enhance non-face region
safety_checker
boolean
True
Safety checker is enabled by default. Un-tick to expose unfiltered results.

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'}