zedge/sdxl-old
Public
456.8K
runs
Run zedge/sdxl-old with an API
Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.
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 |
|---|---|---|---|
| prompt |
string
|
a digital conceptual art illustration, cyberpunk anime digital art, cyberpunk illustration, cyberpunk art station aesthetics, cyberpunk art style, detailed cyberpunk illustration, Digital Conceptual Art Illustration, detailed illustration of game art, detailed digital conceptual art, cinematic lighting
|
User defined prompt
|
| negative_prompt |
string
|
lowres, low resolution, bad quality, jpg artifacts, ugly, deformed, noisy, blurry, noise, low detail, nsfw, topless, see-through, revealing clothing, wrong gender, wrong ethnicity, wrong hair, wrong age, mask, face mask, helmet, face not visible
|
Input Negative Prompt
|
| width |
integer
|
768
|
Width of output image
|
| height |
integer
|
768
|
Height of output image
|
| megapixel_count |
number
|
0.85
Min: 0.5 Max: 3 |
megapixel count for image resizing. 1000x1000 resolution is equal to 1 megapixel
|
| num_outputs |
integer
|
4
Min: 1 Max: 4 |
Number of images to output
|
| num_inference_steps |
integer
|
10
Min: 1 Max: 500 |
Number of denoising steps
|
| strength |
number
|
1
Max: 1 |
How much to transform the reference image (applies to image_to_image model only)
|
| guidance_scale |
number
|
15
Min: 1 Max: 20 |
Scale for classifier-free guidance
|
| scheduler |
None
|
K_EULER
|
Choose a scheduler
|
| seed |
integer
|
-1
|
Seed value. Leave empty to randomize
|
| model_name |
None
|
SDXL_REALVISXL_V4_TEXT_TO_IMAGE
|
Choose a model
|
| use_clip_interrogator |
boolean
|
False
|
Use CLIP interrogator to describe image in img2img flow
|
| verbose |
boolean
|
False
|
Print detailed timing information
|
| return_timing |
boolean
|
False
|
Return timing information in the output
|
| use_compel |
boolean
|
True
|
Use Compel for prompt processing
|
| warm_delay |
integer
|
-1
|
Parameter for warming the model. If set, returns empty dict after specified seconds
|
{
"type": "object",
"title": "Input",
"properties": {
"seed": {
"type": "integer",
"title": "Seed",
"default": -1,
"x-order": 10,
"description": "Seed value. Leave empty to randomize"
},
"width": {
"type": "integer",
"title": "Width",
"default": 768,
"x-order": 2,
"description": "Width of output image"
},
"height": {
"type": "integer",
"title": "Height",
"default": 768,
"x-order": 3,
"description": "Height of output image"
},
"prompt": {
"type": "string",
"title": "Prompt",
"default": "a digital conceptual art illustration, cyberpunk anime digital art, cyberpunk illustration, cyberpunk art station aesthetics, cyberpunk art style, detailed cyberpunk illustration, Digital Conceptual Art Illustration, detailed illustration of game art, detailed digital conceptual art, cinematic lighting",
"x-order": 0,
"description": "User defined prompt"
},
"verbose": {
"type": "boolean",
"title": "Verbose",
"default": false,
"x-order": 13,
"description": "Print detailed timing information"
},
"strength": {
"type": "number",
"title": "Strength",
"default": 1,
"maximum": 1,
"minimum": 0,
"x-order": 7,
"description": "How much to transform the reference image (applies to image_to_image model only)"
},
"scheduler": {
"enum": [
"DDIM",
"K_EULER",
"DPMSolverMultistep",
"K_EULER_ANCESTRAL",
"PNDM",
"KLMS"
],
"type": "string",
"title": "scheduler",
"description": "Choose a scheduler",
"default": "K_EULER",
"x-order": 9
},
"model_name": {
"enum": [
"SDXL_REALVISXL_V4_TEXT_TO_IMAGE",
"SDXL_REALVISXL_V4_IMAGE_TO_IMAGE"
],
"type": "string",
"title": "model_name",
"description": "Choose a model",
"default": "SDXL_REALVISXL_V4_TEXT_TO_IMAGE",
"x-order": 11
},
"use_compel": {
"type": "boolean",
"title": "Use Compel",
"default": true,
"x-order": 15,
"description": "Use Compel for prompt processing"
},
"warm_delay": {
"type": "integer",
"title": "Warm Delay",
"default": -1,
"x-order": 16,
"description": "Parameter for warming the model. If set, returns empty dict after specified seconds"
},
"num_outputs": {
"type": "integer",
"title": "Num Outputs",
"default": 4,
"maximum": 4,
"minimum": 1,
"x-order": 5,
"description": "Number of images to output"
},
"return_timing": {
"type": "boolean",
"title": "Return Timing",
"default": false,
"x-order": 14,
"description": "Return timing information in the output"
},
"guidance_scale": {
"type": "number",
"title": "Guidance Scale",
"default": 15,
"maximum": 20,
"minimum": 1,
"x-order": 8,
"description": "Scale for classifier-free guidance"
},
"megapixel_count": {
"type": "number",
"title": "Megapixel Count",
"default": 0.85,
"maximum": 3,
"minimum": 0.5,
"x-order": 4,
"description": "megapixel count for image resizing. 1000x1000 resolution is equal to 1 megapixel"
},
"negative_prompt": {
"type": "string",
"title": "Negative Prompt",
"default": "lowres, low resolution, bad quality, jpg artifacts, ugly, deformed, noisy, blurry, noise, low detail, nsfw, topless, see-through, revealing clothing, wrong gender, wrong ethnicity, wrong hair, wrong age, mask, face mask, helmet, face not visible",
"x-order": 1,
"description": "Input Negative Prompt"
},
"num_inference_steps": {
"type": "integer",
"title": "Num Inference Steps",
"default": 10,
"maximum": 500,
"minimum": 1,
"x-order": 6,
"description": "Number of denoising steps"
},
"use_clip_interrogator": {
"type": "boolean",
"title": "Use Clip Interrogator",
"default": false,
"x-order": 12,
"description": "Use CLIP interrogator to describe image in img2img flow"
}
}
}
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{
"type": "object",
"title": "Output",
"additionalProperties": true
}