fofr/latent-consistency-model

Super-fast, 0.6s per image. LCM with img2img, large batching and canny controlnet

Capabilities

1:14:33:416:99:162:33:2Reference ImagesSeed

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`archive_outputs`	boolean	Option to archive the output images	`false`	—
`canny_high_threshold`	number	Canny high threshold	`200`	min: 1, max: 255
`canny_low_threshold`	number	Canny low threshold	`100`	min: 1, max: 255
`control_guidance_end`	number	Controlnet end	`1`	min: 0, max: 1
`control_guidance_start`	number	Controlnet start	`0`	min: 0, max: 1
`control_image`	string(uri)	Image for controlnet conditioning	`—`	—
`controlnet_conditioning_scale`	number	Controlnet conditioning scale	`2`	min: 0.1, max: 4
`disable_safety_checker`	boolean	Disable safety checker for generated images. This feature is only available through the API	`false`	—
`guidance_scale`	number	Scale for classifier-free guidance	`8`	min: 1, max: 20
`height`	integer	Height of output image. Lower if out of memory	`768`	—
`image`	string(uri)	Input image for img2img	`—`	—
`lcm_origin_steps`	integer	—	`50`	min: 1
`num_images`	integer	Number of images per prompt	`1`	min: 1, max: 50
`num_inference_steps`	integer	Number of denoising steps. Recommend 1 to 8 steps.	`8`	min: 1, max: 50
`prompt`	string	For multiple prompts, enter each on a new line.	`"Self-portrait oil painting, a beautiful cyborg with golden hair, 8k"`	—
`prompt_strength`	number	Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image	`0.8`	min: 0, max: 1
`seed`	integer	Random seed. Leave blank to randomize the seed	`—`	—
`sizing_strategy`	string	Decide how to resize images – use width/height, resize based on input image or control image	`"width/height"`	width/heightinput_imagecontrol_image
`width`	integer	Width of output image. Lower if out of memory	`768`	—

archive_outputsboolean

Option to archive the output images

Default: false

canny_high_thresholdnumber

Canny high threshold

Default: 200min: 1, max: 255

canny_low_thresholdnumber

Canny low threshold

Default: 100min: 1, max: 255

control_guidance_endnumber

Controlnet end

Default: 1min: 0, max: 1

control_guidance_startnumber

Controlnet start

Default: 0min: 0, max: 1

control_imagestring

Image for controlnet conditioning

controlnet_conditioning_scalenumber

Controlnet conditioning scale

Default: 2min: 0.1, max: 4

disable_safety_checkerboolean

Disable safety checker for generated images. This feature is only available through the API

Default: false

guidance_scalenumber

Scale for classifier-free guidance

Default: 8min: 1, max: 20

heightinteger

Height of output image. Lower if out of memory

Default: 768

imagestring

Input image for img2img

lcm_origin_stepsinteger

Default: 50min: 1

num_imagesinteger

Number of images per prompt

Default: 1min: 1, max: 50

num_inference_stepsinteger

Number of denoising steps. Recommend 1 to 8 steps.

Default: 8min: 1, max: 50

promptstring

For multiple prompts, enter each on a new line.

Default: "Self-portrait oil painting, a beautiful cyborg with golden hair, 8k"

prompt_strengthnumber

Prompt strength when using img2img. 1.0 corresponds to full destruction of information in image

Default: 0.8min: 0, max: 1

seedinteger

Random seed. Leave blank to randomize the seed

sizing_strategystring

Decide how to resize images – use width/height, resize based on input image or control image

Default: "width/height"

width/heightinput_imagecontrol_image

widthinteger

Width of output image. Lower if out of memory

Default: 768

Version: 683d19dc312fUpdated: 7/25/20261.5M runs