usamaehsan / flux-multi-controlnet

Fast FLUX DEV -> Flux Controlnet Canny, Controlnet Depth , Controlnet Line Art, Controlnet Upscaler - You can use just one controlnet or All - LORAs: HyperFlex LoRA , Add Details LoRA , Realism LoRA

  • Public
  • 158.9K runs
  • A100 (80GB)
Iterate in playground

Input

string
Shift + Return to add a new line

The text prompt that guides image generation

Default: "A girl in city, 25 years old, cool, futuristic style"

string
Shift + Return to add a new line

The negative text prompt that guides image generation

Default: "A girl in city, 25 years old, cool, futuristic style"

file

Input image for Canny ControlNet

file

Input image for Depth ControlNet

file

Input image for Lineart ControlNet

file

Input image for Upscaler ControlNet

string

Detector type to use with Canny ControlNet

Default: "canny"

string

Detector type to use with Depth ControlNet

Default: "depth"

string

Detector type to use with Lineart ControlNet

Default: "lineart"

number
(minimum: 0, maximum: 2)

Strength for Canny ControlNet

Default: 0.6

number
(minimum: 0, maximum: 2)

Strength for Depth ControlNet

Default: 0.6

number
(minimum: 0, maximum: 2)

Strength for Lineart ControlNet

Default: 0.6

number
(minimum: 0, maximum: 2)

Strength for Upscaler ControlNet

Default: 0.6

number
(minimum: 0, maximum: 50)

Controls how closely the image follows the prompt

Default: 3.5

integer
(minimum: 1, maximum: 50)

Number of denoising steps

Default: 8

integer

Random seed for reproducibility

number
(maximum: 1)

Weight of the HyperFlex LoRA adaptation

Default: 0.125

number
(maximum: 1)

Weight of the Add Details LoRA adaptation

Default: 0

number
(maximum: 1)

Weight of the Realism LoRA adaptation

Default: 0

number
(minimum: 0, maximum: 1)

Weight of visual-identity-design LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of couple-profile LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of home-decoration LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of portrait-illustration LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of portrait-photography LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of ppt-templates LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of sparklers-visual-effect LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of sparklers-visual-effect LoRA

Default: 0

number
(minimum: 0, maximum: 1)

Weight of sparklers-visual-effect LoRA

Default: 0

file

Input image for inpainting

file

Mask image for inpainting (white pixels will be repainted)

number
(minimum: 0, maximum: 10)

Strength of inpainting (how much to repaint masked area)

Default: 0.6

integer

Size of padding when cropping the mask

number
(minimum: 0, maximum: 1)

Percentage of steps at which ControlNet starts applying

Default: 0

number
(minimum: 0, maximum: 1)

Percentage of steps at which ControlNet stops applying

Default: 1

boolean

Whether to use weighted text embeddings

Default: false

file

Reference image for IP Adapter

number
(minimum: 0, maximum: 10)

Scale of influence for IP Adapter

Default: 0.5

file

Second reference image for IP Adapter

number
(minimum: 0, maximum: 10)

Scale of influence for second IP Adapter image

Default: 0.5

file

Third reference image for IP Adapter

number
(minimum: 0, maximum: 10)

Scale of influence for third IP Adapter image

Default: 0.5

file

Negative reference image for IP Adapter

number
(minimum: 0, maximum: 10)

Scale of influence for negative IP Adapter image

Default: 0.5

boolean

Whether to use automatic masking

Default: false

string
Shift + Return to add a new line

Text prompt for automatic object detection and masking

number
(minimum: 0, maximum: 1)

Box threshold for automatic masking

Default: 0.3

number
(minimum: 0, maximum: 1)

Text threshold for automatic masking

Default: 0.25

boolean

Whether to use background mask instead of object mask

Default: false

Including widthh and 1 more...

Output

output
Generated in

This output was created using a different version of the model, usamaehsan/flux-multi-controlnet:85977e12.

Run time and cost

This model costs approximately $0.017 to run on Replicate, or 58 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 A100 (80GB) GPU hardware. Predictions typically complete within 13 seconds. The predict time for this model varies significantly based on the inputs.

Readme

This model doesn't have a readme.