fofr/sdxl-multi-controlnet-lora

Multi-controlnet, lora loading, img2img, inpainting

Capabilities

1:14:33:416:99:162:33:2Reference ImagesNegative PromptSeed

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`apply_watermark`	boolean	Applies a watermark to enable determining if an image is generated in downstream applications. If you have other provisions for generating or deploying images safely, you can use this to disable watermarking.	`true`	—
`controlnet_1`	string	Controlnet	`"none"`	noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose
`controlnet_1_conditioning_scale`	number	How strong the controlnet conditioning is	`0.75`	min: 0, max: 4
`controlnet_1_end`	number	When controlnet conditioning ends	`1`	min: 0, max: 1
`controlnet_1_image`	string(uri)	Input image for first controlnet	`—`	—
`controlnet_1_start`	number	When controlnet conditioning starts	`0`	min: 0, max: 1
`controlnet_2`	string	Controlnet	`"none"`	noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose
`controlnet_2_conditioning_scale`	number	How strong the controlnet conditioning is	`0.75`	min: 0, max: 4
`controlnet_2_end`	number	When controlnet conditioning ends	`1`	min: 0, max: 1
`controlnet_2_image`	string(uri)	Input image for second controlnet	`—`	—
`controlnet_2_start`	number	When controlnet conditioning starts	`0`	min: 0, max: 1
`controlnet_3`	string	Controlnet	`"none"`	noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose
`controlnet_3_conditioning_scale`	number	How strong the controlnet conditioning is	`0.75`	min: 0, max: 4
`controlnet_3_end`	number	When controlnet conditioning ends	`1`	min: 0, max: 1
`controlnet_3_image`	string(uri)	Input image for third controlnet	`—`	—
`controlnet_3_start`	number	When controlnet conditioning starts	`0`	min: 0, max: 1
`disable_safety_checker`	boolean	Disable safety checker for generated images. This feature is only available through the API. See [https://replicate.com/docs/how-does-replicate-work#safety](https://replicate.com/docs/how-does-replicate-work#safety)	`false`	—
`guidance_scale`	number	Scale for classifier-free guidance	`7.5`	min: 1, max: 50
`height`	integer	Height of output image	`768`	—
`image`	string(uri)	Input image for img2img or inpaint mode	`—`	—
`lora_scale`	number	LoRA additive scale. Only applicable on trained models.	`0.6`	min: 0, max: 1
`lora_weights`	string	Replicate LoRA weights to use. Leave blank to use the default weights.	`—`	—
`mask`	string(uri)	Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.	`—`	—
`negative_prompt`	string	Negative Prompt	`""`	—
`num_inference_steps`	integer	Number of denoising steps	`30`	min: 1, max: 500
`num_outputs`	integer	Number of images to output	`1`	min: 1, max: 4
`prompt`	string	Input prompt	`"An astronaut riding a rainbow unicorn"`	—
`prompt_strength`	number	Prompt strength when using img2img / inpaint. 1.0 corresponds to full destruction of information in image	`0.8`	min: 0, max: 1
`refine`	string	Which refine style to use	`"no_refiner"`	no_refinerbase_image_refiner
`refine_steps`	integer	For base_image_refiner, the number of steps to refine, defaults to num_inference_steps	`—`	—
`scheduler`	string	scheduler	`"K_EULER"`	DDIMDPMSolverMultistepHeunDiscreteKarrasDPMK_EULER_ANCESTRALK_EULERPNDM
`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_imagecontrolnet_1_imagecontrolnet_2_imagecontrolnet_3_imagemask_image
`width`	integer	Width of output image	`768`	—

apply_watermarkboolean

Applies a watermark to enable determining if an image is generated in downstream applications. If you have other provisions for generating or deploying images safely, you can use this to disable watermarking.

Default: true

controlnet_1string

Controlnet

Default: "none"

noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose

controlnet_1_conditioning_scalenumber

How strong the controlnet conditioning is

Default: 0.75min: 0, max: 4

controlnet_1_endnumber

When controlnet conditioning ends

Default: 1min: 0, max: 1

controlnet_1_imagestring

Input image for first controlnet

controlnet_1_startnumber

When controlnet conditioning starts

Default: 0min: 0, max: 1

controlnet_2string

Controlnet

Default: "none"

noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose

controlnet_2_conditioning_scalenumber

How strong the controlnet conditioning is

Default: 0.75min: 0, max: 4

controlnet_2_endnumber

When controlnet conditioning ends

Default: 1min: 0, max: 1

controlnet_2_imagestring

Input image for second controlnet

controlnet_2_startnumber

When controlnet conditioning starts

Default: 0min: 0, max: 1

controlnet_3string

Controlnet

Default: "none"

noneedge_cannyillusiondepth_leresdepth_midassoft_edge_pidisoft_edge_hedlineartlineart_animeopenpose

controlnet_3_conditioning_scalenumber

How strong the controlnet conditioning is

Default: 0.75min: 0, max: 4

controlnet_3_endnumber

When controlnet conditioning ends

Default: 1min: 0, max: 1

controlnet_3_imagestring

Input image for third controlnet

controlnet_3_startnumber

When controlnet conditioning starts

Default: 0min: 0, max: 1

disable_safety_checkerboolean

Disable safety checker for generated images. This feature is only available through the API. See [https://replicate.com/docs/how-does-replicate-work#safety](https://replicate.com/docs/how-does-replicate-work#safety)

Default: false

guidance_scalenumber

Scale for classifier-free guidance

Default: 7.5min: 1, max: 50

heightinteger

Height of output image

Default: 768

imagestring

Input image for img2img or inpaint mode

lora_scalenumber

LoRA additive scale. Only applicable on trained models.

Default: 0.6min: 0, max: 1

lora_weightsstring

Replicate LoRA weights to use. Leave blank to use the default weights.

maskstring

Input mask for inpaint mode. Black areas will be preserved, white areas will be inpainted.

negative_promptstring

Negative Prompt

Default: ""

num_inference_stepsinteger

Number of denoising steps

Default: 30min: 1, max: 500

num_outputsinteger

Number of images to output

Default: 1min: 1, max: 4

promptstring

Input prompt

Default: "An astronaut riding a rainbow unicorn"

prompt_strengthnumber

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

Default: 0.8min: 0, max: 1

refinestring

Which refine style to use

Default: "no_refiner"

no_refinerbase_image_refiner

refine_stepsinteger

For base_image_refiner, the number of steps to refine, defaults to num_inference_steps

schedulerstring

scheduler

Default: "K_EULER"

DDIMDPMSolverMultistepHeunDiscreteKarrasDPMK_EULER_ANCESTRALK_EULERPNDM

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_imagecontrolnet_1_imagecontrolnet_2_imagecontrolnet_3_imagemask_image

widthinteger

Width of output image

Default: 768

Version: 89eb212b3d13Updated: 7/25/2026217.9K runs