You're looking at a specific version of this model. Jump to the model overview.
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 |
|---|---|---|---|
| input_image |
string
|
Input face image
|
|
| prompt |
string
|
vector art, vibrant neon colors, retro 80s Miami aesthetic, bold outlines, flat shading, urban cityscape background, palm trees, sunset sky, comic book style, high contrast, saturated colors
|
Input prompt
|
| negative_prompt |
string
|
nsfw, nude, watermark, text, logo, signature, jpeg artifacts, blurry, out of focus, low quality, disfigured, deformed, mutated, ugly
|
Input Negative Prompt
|
| sdxl_weights |
None
|
RealVisXL_V5.0
|
Pick which base weights you want to use
|
| scheduler |
None
|
EulerDiscreteScheduler
|
Scheduler
|
| num_inference_steps |
integer
|
30
Min: 1 Max: 500 |
Number of denoising steps
|
| guidance_scale |
number
|
7.5
Min: 1 Max: 50 |
Scale for classifier-free guidance
|
| ip_adapter_scale |
number
|
0.8
Max: 1.5 |
ONLY FOR INSTANTID: Scale for image adapter strength (for detail)
|
| identitynet_strength_ratio |
number
|
0.8
Max: 1.5 |
ONLY FOR INSTANTID: Scale for IdentityNet strength (for fidelity)
|
| instantid_pose_strength |
number
|
0
Max: 1 |
ONLY FOR INSTANTID: Openpose ControlNet strength, effective only if > 0
|
| instantid_canny_strength |
number
|
0.3
Max: 1 |
ONLY FOR INSTANTID: Canny ControlNet strength, effective only if > 0
|
| instantid_depth_strength |
number
|
0.8
Max: 1 |
ONLY FOR INSTANTID: Depth ControlNet strength, effective only if > 0
|
| img2img_canny_strength |
number
|
0
Max: 1 |
ONLY FOR 0 FACES: Canny ControlNet strength, effective only if > 0
|
| img2img_depth_strength |
number
|
0.7
Max: 1 |
ONLY FOR 0 FACES: Depth ControlNet strength, effective only if > 0
|
| enable_lcm |
boolean
|
False
|
Enable Fast Inference with LCM (Latent Consistency Models) - speeds up inference steps, trade-off is the quality of the generated image. Performs better with close-up portrait face images
|
| lcm_num_inference_steps |
integer
|
5
Min: 1 Max: 10 |
Only used when `enable_lcm` is set to True, Number of denoising steps when using LCM
|
| lcm_guidance_scale |
number
|
1.5
Min: 1 Max: 20 |
Only used when `enable_lcm` is set to True, Scale for classifier-free guidance when using LCM
|
| enhance_nonface_region |
boolean
|
True
|
ONLY FOR INSTANTID: Enhance non-face region
|
| seed |
integer
|
Random seed. Leave blank to randomize the seed
|
|
| num_outputs |
integer
|
1
Min: 1 Max: 4 |
Number of images to output
|
| input_zip |
string
|
ZIP file containing input images.
|
|
| megapixel_count |
number
|
1
|
Megapixel count for image downscaling. 1024x1024 resolution is equal to 1 megapixel
|
| disable_nsfw_checker |
boolean
|
False
|
Disable safety checker for generated images.
|
| verbose |
boolean
|
False
|
Print detailed timing information
|
| debug_images |
boolean
|
False
|
(PARAMETER ONLY RELEVANT IN DEVELOPMENT) Save debug images
|
| force_clip |
boolean
|
False
|
(DEPRECATED) Force using CLIP captioning regardless of face detection
|
| warm_delay |
integer
|
Parameter for warming the model. If set, returns empty dict after specified seconds
|
Output schema
The shape of the response you’ll get when you run this model with an API.
{'title': 'Output', 'type': 'object'}