underthestar2021 / mxcfsuper-public

  • Public
  • 42.3K runs
  • L40S

Input

file

Input face image

string
Shift + Return to add a new line

备注项,随便填

string
Shift + Return to add a new line

Input prompt

Default: "a photo of an astronaut riding a horse on mars"

string
Shift + Return to add a new line

Specify things to not see in the output

number
(minimum: 0, maximum: 1)

Prompt strength when providing the image. 1.0 corresponds to full destruction of information in init image

Default: 0.7

string

画图类型,文生图还是图生图。不可以选空

Default: "txt2img"

integer
(minimum: 1)

Default: 2

integer

Random seed. Leave blank to randomize the seed

integer
(minimum: 1, maximum: 4)

Number of images to output.

Default: 1

integer
(minimum: 1, maximum: 500)

Number of denoising steps

Default: 26

number
(minimum: 1, maximum: 20)

Scale for classifier-free guidance

Default: 7.5

integer
(minimum: 1, maximum: 1024)

Width of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

Default: 512

integer
(minimum: 1, maximum: 1024)

Height of output image. Maximum size is 1024x768 or 768x1024 because of memory limits

Default: 720

string

Choose a scheduler.

Default: "DPM++ SDE Karras"

boolean

是否启用Controlnet

Default: true

string

An enumeration.

Default: "canny"

number
(minimum: 0, maximum: 2)

Control Weight

Default: 1

integer
(minimum: 1, maximum: 255)

Canny line detection low threshold

Default: 100

integer
(minimum: 1, maximum: 255)

Canny line detection high threshold

Default: 200

number
(minimum: 0, maximum: 1)

The percentage of total steps at which the controlnet starts applying

Default: 0

number
(minimum: 0, maximum: 1)

The percentage of total steps at which the controlnet stops applying

Default: 1

string

An enumeration.

string
Shift + Return to add a new line

测试lora时使用;lora模型的下载地址,如果填写会复盖掉lora_model参数

number

lora权重,区间0-1. A scale value of 0 is the same as not using your LoRA weights and you’re only using the base model weights, and a scale value of 1 means you’re only using the fully fine-tuned LoRA weights.

Default: 0.8

boolean

是否使用swap_face(roop的换脸功能),默认True

Default: true

boolean

是否使用restore_face,默认True

Default: true

integer
(minimum: 1, maximum: 4)

Restore face upscaling

Default: 1

boolean

Restore face upsampling

Default: true

boolean

Restore face background enhance

Default: true

number
(minimum: 0, maximum: 1)

Codeformer fidelity

Default: 0.7

boolean

是否使用face editor,默认False

Default: false

number
(minimum: 0.1, maximum: 0.8)

Step 0.05, face editor, denoising strength for face images, sdwebui:(3) Recreate the Faces

Default: 0.4

number
(minimum: 0, maximum: 1)

Step 0.05, face editor, denoising strength for the entire image, sdwebui:(5) Blend the entire image

Default: 0.05

integer

Face editor mask size

Default: 6

integer

Face editor mask blur

Default: 12

number
(minimum: 0.7, maximum: 1)

Confidence threshold for face detection. Set a lower value if you want to detect more faces.

Default: 0.97

string

fill - The image is resized to fill the given dimension. cover - The image keeps its aspect ratio and fills the given dimension. The image will be clipped to fit. crop - The image keeps its aspect ratio and scales to the target size.

Default: "fill"

string
Shift + Return to add a new line

Unique identifier for this run. Will be used in callback.

string
Shift + Return to add a new line

Task ID for this run. Will be used in callback.

integer

api version

Default: 4

string
Shift + Return to add a new line

Process callback URL for this .

string
Shift + Return to add a new line

Complete callback URL for this run.

integer
(minimum: 1, maximum: 100)

Callback URL for this run.

Default: 4

string
Shift + Return to add a new line

random seed

Default: "ai_random"

boolean

Show system info

Default: false

Output

No output yet! Press "Submit" to start a prediction.

Run time and cost

This model costs approximately $0.078 to run on Replicate, or 12 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 81 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.