usamaehsan/controlnet-x-ip-adapter-realistic-vision-v5

OfficialView on Replicate →

Inpainting || multi-controlnet || single-controlnet || ip-adapter || ip adapter face || ip adapter plus || No ip adapter

Capabilities

Negative PromptSeed

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`prompt`*	string	Prompt - using compel, use +++ to increase words weight:: doc: https://github.com/damian0815/compel/tree/main/doc \|\| https://invoke-ai.github.io/InvokeAI/features/PROMPTS/#attention-weighting	`—`	—
`add_more_detail_lora_scale`	number	Scale/ weight of more_details lora, more scale = more details, disabled on 0	`0.5`	—
`brightness_conditioning_scale`	number	Conditioning scale for brightness controlnet	`1`	—
`brightness_image`	string(uri)	Control image for brightness controlnet	`—`	—
`color_temprature_slider_lora_weight`	number	disabled on 0	`0`	—
`detail_tweaker_lora_weight`	number	disabled on 0	`0`	—
`disable_safety_check`	boolean	Disable safety check. Use at your own risk!	`false`	—
`epi_noise_offset_lora_weight`	number	disabled on 0	`0`	—
`eta`	number	Controls the amount of noise that is added to the input data during the denoising diffusion process. Higher value -> more noise	`0`	—
`film_grain_lora_weight`	number	disabled on 0	`0`	—
`guess_mode`	boolean	In this mode, the ControlNet encoder will try best to recognize the content of the input image even if you remove all prompts. The `guidance_scale` between 3.0 and 5.0 is recommended.	`false`	—
`guidance_scale`	number	Scale for classifier-free guidance	`7`	min: 0.1, max: 30
`img2img_image`	string(uri)	Image2image image	`—`	—
`img2img_strength`	number	img2img strength, does not work when inpainting image is given, 0.1-same image, 0.99-complete destruction of image	`0.5`	—
`inpainting_conditioning_scale`	number	Conditioning scale for inpaint controlnet	`1`	—
`inpainting_image`	string(uri)	Control image for inpainting controlnet	`—`	—
`inpainting_strength`	number	inpainting strength	`1`	—
`int_kwargs`	string	—	`""`	—
`ip_adapter_ckpt`	string	IP Adapter checkpoint	`"ip-adapter_sd15.bin"`	ip-adapter_sd15.binip-adapter-plus_sd15.binip-adapter-plus-face_sd15.bin
`ip_adapter_image`	string(uri)	IP Adapter image	`—`	—
`ip_adapter_weight`	number	IP Adapter weight	`1`	—
`lineart_conditioning_scale`	number	Conditioning scale for canny controlnet	`1`	—
`lineart_image`	string(uri)	Control image for canny controlnet	`—`	—
`mask_image`	string(uri)	mask image for inpainting controlnet	`—`	—
`max_height`	integer	Max height/Resolution of image	`512`	—
`max_width`	integer	Max width/Resolution of image	`512`	—
`negative_auto_mask_text`	string	// seperated list of objects you dont want to mask - 'hairs // eyes // cloth'	`—`	—
`negative_prompt`	string	Negative prompt - using compel, use +++ to increase words weight//// negative-embeddings available ///// FastNegativeV2 , boring_e621_v4 , verybadimagenegative_v1 \|\| to use them, write their keyword in negative prompt	`"Longbody, lowres, bad anatomy, bad hands, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality"`	—
`num_inference_steps`	integer	Steps to run denoising	`20`	—
`num_outputs`	integer	Number of images to generate	`1`	min: 1, max: 10
`positive_auto_mask_text`	string	// seperated list of objects for mask, AI will auto create mask of these objects, if mask text is given, mask image will not work - 'hairs // eyes // cloth'	`—`	—
`scheduler`	string	Choose a scheduler.	`"DDIM"`	DDIMDPMSolverMultistepHeunDiscreteK_EULER_ANCESTRALK_EULERKLMSPNDMUniPCMultistepKDPM2DiscreteSchedulerKDPM2AncestralDiscreteSchedulerDDPMSchedulerDEISMultistepScheduler
`scribble_conditioning_scale`	number	Conditioning scale for scribble controlnet	`1`	—
`scribble_image`	string(uri)	Control image for scribble controlnet	`—`	—
`seed`	integer	Seed	`—`	—
`sorted_controlnets`	string	Comma seperated string of controlnet names, list of names: tile, inpainting, lineart,depth ,scribble , brightness /// example value: tile, inpainting, lineart	`"lineart, tile, inpainting"`	—
`tile_conditioning_scale`	number	Conditioning scale for tile controlnet	`1`	—
`tile_image`	string(uri)	Control image for tile controlnet	`—`	—

promptrequiredstring

Prompt - using compel, use +++ to increase words weight:: doc: https://github.com/damian0815/compel/tree/main/doc || https://invoke-ai.github.io/InvokeAI/features/PROMPTS/#attention-weighting

add_more_detail_lora_scalenumber

Scale/ weight of more_details lora, more scale = more details, disabled on 0

Default: 0.5

brightness_conditioning_scalenumber

Conditioning scale for brightness controlnet

Default: 1

brightness_imagestring

Control image for brightness controlnet

color_temprature_slider_lora_weightnumber

disabled on 0

Default: 0

detail_tweaker_lora_weightnumber

disabled on 0

Default: 0

disable_safety_checkboolean

Disable safety check. Use at your own risk!

Default: false

epi_noise_offset_lora_weightnumber

disabled on 0

Default: 0

etanumber

Controls the amount of noise that is added to the input data during the denoising diffusion process. Higher value -> more noise

Default: 0

film_grain_lora_weightnumber

disabled on 0

Default: 0

guess_modeboolean

In this mode, the ControlNet encoder will try best to recognize the content of the input image even if you remove all prompts. The `guidance_scale` between 3.0 and 5.0 is recommended.

Default: false

guidance_scalenumber

Scale for classifier-free guidance

Default: 7min: 0.1, max: 30

img2img_imagestring

Image2image image

img2img_strengthnumber

img2img strength, does not work when inpainting image is given, 0.1-same image, 0.99-complete destruction of image

Default: 0.5

inpainting_conditioning_scalenumber

Conditioning scale for inpaint controlnet

Default: 1

inpainting_imagestring

Control image for inpainting controlnet

inpainting_strengthnumber

inpainting strength

Default: 1

int_kwargsstring

Default: ""

ip_adapter_ckptstring

IP Adapter checkpoint

Default: "ip-adapter_sd15.bin"

ip-adapter_sd15.binip-adapter-plus_sd15.binip-adapter-plus-face_sd15.bin

ip_adapter_imagestring

IP Adapter image

ip_adapter_weightnumber

IP Adapter weight

Default: 1

lineart_conditioning_scalenumber

Conditioning scale for canny controlnet

Default: 1

lineart_imagestring

Control image for canny controlnet

mask_imagestring

mask image for inpainting controlnet

max_heightinteger

Max height/Resolution of image

Default: 512

max_widthinteger

Max width/Resolution of image

Default: 512

negative_auto_mask_textstring

// seperated list of objects you dont want to mask - 'hairs // eyes // cloth'

negative_promptstring

Negative prompt - using compel, use +++ to increase words weight//// negative-embeddings available ///// FastNegativeV2 , boring_e621_v4 , verybadimagenegative_v1 || to use them, write their keyword in negative prompt

Default:

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

num_inference_stepsinteger

Steps to run denoising

Default: 20

num_outputsinteger

Number of images to generate

Default: 1min: 1, max: 10

positive_auto_mask_textstring

// seperated list of objects for mask, AI will auto create mask of these objects, if mask text is given, mask image will not work - 'hairs // eyes // cloth'

schedulerstring

Choose a scheduler.

Default: "DDIM"

DDIMDPMSolverMultistepHeunDiscreteK_EULER_ANCESTRALK_EULERKLMSPNDMUniPCMultistepKDPM2DiscreteSchedulerKDPM2AncestralDiscreteSchedulerDDPMSchedulerDEISMultistepScheduler

scribble_conditioning_scalenumber

Conditioning scale for scribble controlnet

Default: 1

scribble_imagestring

Control image for scribble controlnet

seedinteger

Seed

sorted_controlnetsstring

Comma seperated string of controlnet names, list of names: tile, inpainting, lineart,depth ,scribble , brightness /// example value: tile, inpainting, lineart

Default: "lineart, tile, inpainting"

tile_conditioning_scalenumber

Conditioning scale for tile controlnet

Default: 1

tile_imagestring

Control image for tile controlnet

Version: 50ac06bb9bcfUpdated: 7/25/2026674.7K runs