jagilley/controlnet-seg

Modify images using semantic segmentation

Capabilities

Reference Images Negative Prompt Seed

Community model (estimated from hardware time)

Name	Type	Description	Default	Constraints
`image` *	string (uri)	Input image	`—`	—
`prompt` *	string	Prompt for the model	`—`	—
`a_prompt`	string	Added Prompt	`"best quality, extremely detailed"`	—
`ddim_steps`	integer	Steps	`20`	—
`detect_resolution`	integer	Resolution for detection (only applicable when model type is 'HED' or 'Segmentation')	`512`	min: 128, max: 1024
`eta`	number	eta (DDIM)	`0`	—
`image_resolution`	string	Image resolution to be generated	`"512"`	256 512 768
`n_prompt`	string	Negative Prompt	`"longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality"`	—
`num_samples`	string	Number of samples (higher values may OOM)	`"1"`	1 4
`scale`	number	Guidance Scale	`9`	min: 0.1, max: 30
`seed`	integer	Seed	`—`	—

image required string

Input image

prompt required string

Prompt for the model

a_prompt string

Added Prompt

Default: "best quality, extremely detailed"

ddim_steps integer

Steps

Default: 20

detect_resolution integer

Resolution for detection (only applicable when model type is 'HED' or 'Segmentation')

Default: 512 min: 128, max: 1024

eta number

eta (DDIM)

Default: 0

image_resolution string

Image resolution to be generated

Default: "512"

256 512 768

n_prompt string

Negative Prompt

Default:

"longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality"

num_samples string

Number of samples (higher values may OOM)

Default: "1"

1 4

scale number

Guidance Scale

Default: 9 min: 0.1, max: 30

seed integer

Seed

Version: f967b165f4cd Updated: 2/26/2026 166.8K runs