mlscade / stable_diffusion_no_ui_control

  • Public
  • 111.9K runs
Run with an API
Playground API Examples README Versions

Run mlscade/stable_diffusion_no_ui_control with an API

Use one of our client libraries to get started quickly. Clicking on a library will take you to the Playground tab where you can tweak different inputs, see the results, and copy the corresponding code to use in your own project.

Input schema

The fields you can use to run this model with an API. If you don't give a value for a field its default value will be used.

Field Type Default value Description
generation_type
string (enum)
txt2img

Options:

txt2img, img2img, postprocess

Generation mode
clip_stop_at_last_layers
integer
2

Min: 1

Max: 12

Clip skip
tiling
boolean
False
Enable tiling
upload_lora
string
Upload LoRA
prompt
string
Prompt
negative_prompt
string
(deformed, distorted, disfigured:1.3), (naked, nude:1.5), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation
Negative prompt
image
string
Input image [img2img or postprocessing modes only]
cfg_scale
number
7
CFG scale
seed
integer
Seed. Leave blank for random generation
width
integer
512
Width
height
integer
512
Height
num_outputs
integer
1

Min: 1

Max: 4

Batch size [Doesn't work in postprocess mode].
sampler
string (enum)
Euler a

Options:

DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC

Sampling method
sampler_steps
integer
20

Min: 1

Max: 50

Sampling steps [Doesn't work in postprocess mode].
hr_enabled
boolean
False
[HR] Enable
denoising_strength
number
0.75
[HR] Denoising strength
hr_scale
number
2
[HR] Upscale by
hr_second_pass_steps
number
0
[HR] Hires steps
hr_upscaler
string (enum)
4x_foolhardy_Remacri

Options:

Latent, Latent (antialiased), Latent (bicubic), Latent (bicubic antialiased), Latent (nearest), Latent (nearest-exact), 4x-AnimeSharp, 4x_foolhardy_Remacri, 8x_NMKD-Superscale_150000_G, Без модели

[HR] Upscaler
hr_resize_x
integer
0
[HR] Resize width to
hr_resize_y
integer
0
[HR] Resize height to
ad_enabled
boolean
False
[adetailer] Enable
ad_model
string (enum)
face_yolov8s

Options:

face_yolov8n, face_yolov8s, hand_yolov8n, person_yolov8n-seg, person_yolov8s-seg, mediapipe_face_full, mediapipe_face_short, mediapipe_face_mesh, mediapipe_face_mesh_eyes_only, deepfashion2_yolov8s-seg, Без модели

[adetailer] ADetailer model
ad_prompt
string
[adetailer] ADetailer prompt
ad_negative_prompt
string
[adetailer] ADetailer negative prompt
ad_use_sampler
boolean
True
[adetailer] Use separate sampler
ad_sampler
string (enum)
Euler a

Options:

DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC

[adetailer] ADetailer sampler
ad_confidence
number
0.3
[adetailer] Detection model confidence threshold
ad_dilate_erode
integer
4
[adetailer] Mask erosion (-) / dilation (+)
ad_mask_blur
integer
4
[adetailer] Inpaint mask blur
ad_mask_merge_mode
string (enum)
None

Options:

None, Merge, Merge and Invert

[adetailer] Mask merge mode
ad_denoising_strength
number
0.4
[adetailer] Inpaint denoising strength
ad_inpaint_only_masked
boolean
True
[adetailer] Inpaint only masked
ad_inpaint_only_masked_padding
integer
32
[adetailer] Inpaint only masked padding, pixels
ad_use_inpaint_width_height
boolean
False
[adetailer] Use separate width/height
ad_inpaint_width
integer
512
[adetailer] Inpaint width
ad_inpaint_height
integer
512
[adetailer] Inpaint height
ad_2_ad_enabled
boolean
False
[adetailer — 2] Enable
ad_2_ad_model
string (enum)
face_yolov8s

Options:

face_yolov8n, face_yolov8s, hand_yolov8n, person_yolov8n-seg, person_yolov8s-seg, mediapipe_face_full, mediapipe_face_short, mediapipe_face_mesh, mediapipe_face_mesh_eyes_only, deepfashion2_yolov8s-seg, Без модели

[adetailer — 2] ADetailer model
ad_2_ad_prompt
string
[adetailer — 2] ADetailer prompt
ad_2_ad_negative_prompt
string
[adetailer — 2] ADetailer negative prompt
ad_2_ad_use_sampler
boolean
True
[adetailer — 2] Use separate sampler
ad_2_ad_sampler
string (enum)
Euler a

Options:

DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC

[adetailer — 2] ADetailer sampler
ad_2_ad_confidence
number
0.3
[adetailer — 2] Detection model confidence threshold
ad_2_ad_dilate_erode
integer
4
[adetailer — 2] Mask erosion (-) / dilation (+)
ad_2_ad_mask_blur
integer
4
[adetailer — 2] Inpaint mask blur
ad_2_ad_mask_merge_mode
string (enum)
None

Options:

None, Merge, Merge and Invert

[adetailer — 2] Mask merge mode
ad_2_ad_denoising_strength
number
0.4
[adetailer — 2] Inpaint denoising strength
ad_2_ad_inpaint_only_masked
boolean
True
[adetailer — 2] Inpaint only masked
ad_2_ad_inpaint_only_masked_padding
integer
32
[adetailer — 2] Inpaint only masked padding, pixels
ad_2_ad_use_inpaint_width_height
boolean
False
[adetailer — 2] Use separate width/height
ad_2_ad_inpaint_width
integer
512
[adetailer — 2] Inpaint width
ad_2_ad_inpaint_height
integer
512
[adetailer — 2] Inpaint height
ad_3_ad_enabled
boolean
False
[adetailer — 3] Enable
ad_3_ad_model
string (enum)
face_yolov8s

Options:

face_yolov8n, face_yolov8s, hand_yolov8n, person_yolov8n-seg, person_yolov8s-seg, mediapipe_face_full, mediapipe_face_short, mediapipe_face_mesh, mediapipe_face_mesh_eyes_only, deepfashion2_yolov8s-seg, Без модели

[adetailer — 3] ADetailer model
ad_3_ad_prompt
string
[adetailer — 3] ADetailer prompt
ad_3_ad_negative_prompt
string
[adetailer — 3] ADetailer negative prompt
ad_3_ad_use_sampler
boolean
True
[adetailer — 3] Use separate sampler
ad_3_ad_sampler
string (enum)
Euler a

Options:

DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC

[adetailer — 3] ADetailer sampler
ad_3_ad_confidence
number
0.3
[adetailer — 3] Detection model confidence threshold
ad_3_ad_dilate_erode
integer
4
[adetailer — 3] Mask erosion (-) / dilation (+)
ad_3_ad_mask_blur
integer
4
[adetailer — 3] Inpaint mask blur
ad_3_ad_mask_merge_mode
string (enum)
None

Options:

None, Merge, Merge and Invert

[adetailer — 3] Mask merge mode
ad_3_ad_denoising_strength
number
0.4
[adetailer — 3] Inpaint denoising strength
ad_3_ad_inpaint_only_masked
boolean
True
[adetailer — 3] Inpaint only masked
ad_3_ad_inpaint_only_masked_padding
integer
32
[adetailer — 3] Inpaint only masked padding, pixels
ad_3_ad_use_inpaint_width_height
boolean
False
[adetailer — 3] Use separate width/height
ad_3_ad_inpaint_width
integer
512
[adetailer — 3] Inpaint width
ad_3_ad_inpaint_height
integer
512
[adetailer — 3] Inpaint height
ad_4_ad_enabled
boolean
False
[adetailer — 4] Enable
ad_4_ad_model
string (enum)
face_yolov8s

Options:

face_yolov8n, face_yolov8s, hand_yolov8n, person_yolov8n-seg, person_yolov8s-seg, mediapipe_face_full, mediapipe_face_short, mediapipe_face_mesh, mediapipe_face_mesh_eyes_only, deepfashion2_yolov8s-seg, Без модели

[adetailer — 4] ADetailer model
ad_4_ad_prompt
string
[adetailer — 4] ADetailer prompt
ad_4_ad_negative_prompt
string
[adetailer — 4] ADetailer negative prompt
ad_4_ad_use_sampler
boolean
True
[adetailer — 4] Use separate sampler
ad_4_ad_sampler
string (enum)
Euler a

Options:

DPM++ 2M Karras, DPM++ SDE Karras, DPM++ 2M SDE Exponential, DPM++ 2M SDE Karras, Euler a, Euler, LMS, Heun, DPM2, DPM2 a, DPM++ 2S a, DPM++ SDE, DPM++ 2M SDE, DPM++ 2M SDE Heun Karras, DPM++ 2M SDE Heun Exponential, DPM++ 3M SDE, DPM++ 3M SDE Karras, DPM++ 3M SDE Exponential, DPM fast, DPM adaptive, LMS Karras, DPM2 Karras, DPM2 a Karras, DPM++ 2S a Karras, Restart, DDIM, PLMS, UniPC

[adetailer — 4] ADetailer sampler
ad_4_ad_confidence
number
0.3
[adetailer — 4] Detection model confidence threshold
ad_4_ad_dilate_erode
integer
4
[adetailer — 4] Mask erosion (-) / dilation (+)
ad_4_ad_mask_blur
integer
4
[adetailer — 4] Inpaint mask blur
ad_4_ad_mask_merge_mode
string (enum)
None

Options:

None, Merge, Merge and Invert

[adetailer — 4] Mask merge mode
ad_4_ad_denoising_strength
number
0.4
[adetailer — 4] Inpaint denoising strength
ad_4_ad_inpaint_only_masked
boolean
True
[adetailer — 4] Inpaint only masked
ad_4_ad_inpaint_only_masked_padding
integer
32
[adetailer — 4] Inpaint only masked padding, pixels
ad_4_ad_use_inpaint_width_height
boolean
False
[adetailer — 4] Use separate width/height
ad_4_ad_inpaint_width
integer
512
[adetailer — 4] Inpaint width
ad_4_ad_inpaint_height
integer
512
[adetailer — 4] Inpaint height
sd_upscaler_enable
boolean
False
[SD upscaler] Enable
sd_upscaler_upscaler
string (enum)
4x_foolhardy_Remacri

Options:

4x-AnimeSharp, 4x_foolhardy_Remacri, 8x_NMKD-Superscale_150000_G, Без модели

[SD upscaler] Upscaler
sd_upscaler_tile_overlap
integer
16
[SD upscaler] Tile overlap
sd_upscaler_scale_factor
number
2
[SD upscaler] Scale Factor
openpose_enabled
boolean
False
[ControlNet — openpose] Enable
openpose_image
string
[ControlNet — openpose] Input image
openpose_use_image_size
boolean
False
[ControlNet — openpose] Use provided image size
openpose_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — openpose] Control Mode
openpose_preprocessor
string (enum)
openpose_full

Options:

openpose_full, dw_openpose_full, openpose, openpose_face, openpose_faceonly, openpose_hand, Без препроцессора

[ControlNet — openpose] Preprocessor
openpose_model
string (enum)
controlnet11Models_openpose

Options:

controlnet11Models_openpose, Без модели

[ControlNet — openpose] Model
openpose_control_weight
number
1

Max: 2

[ControlNet — openpose] Control Weight
openpose_guidance_start
number
0

Max: 1

[ControlNet — openpose] Starting Control Step
openpose_guidance_end
number
1

Max: 1

[ControlNet — openpose] Ending Control Step
openpose_preprocessor_resolution
integer
512
[ControlNet — openpose] Preprocessor Resolution
scribble_enabled
boolean
False
[ControlNet — scribble] Enable
scribble_image
string
[ControlNet — scribble] Input image
scribble_use_image_size
boolean
False
[ControlNet — scribble] Use provided image size
scribble_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — scribble] Control Mode
scribble_preprocessor
string (enum)
t2ia_sketch_pidi

Options:

scribble_xdog, t2ia_sketch_pidi, Без препроцессора

[ControlNet — scribble] Preprocessor
scribble_model
string (enum)
controlnet11Models_scribble

Options:

controlnet11Models_scribble, Без модели

[ControlNet — scribble] Model
scribble_control_weight
number
1

Max: 2

[ControlNet — scribble] Control Weight
scribble_guidance_start
number
0

Max: 1

[ControlNet — scribble] Starting Control Step
scribble_guidance_end
number
1

Max: 1

[ControlNet — scribble] Ending Control Step
scribble_preprocessor_resolution
integer
512

Min: 64

Max: 2048

[ControlNet — scribble] Preprocessor Resolution
scribble_xdog_threshold
integer
32

Min: 1

Max: 64

[ControlNet — scribble] XDoG Threshold [only scribble_xdog preprocessor]
tile_blur_enabled
boolean
False
[ControlNet — tile/blur] Enable
tile_blur_image
string
[ControlNet — tile/blur] Input image
tile_blur_use_image_size
boolean
False
[ControlNet — tile/blur] Use provided image size
tile_blur_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — tile/blur] Control Mode
tile_blur_preprocessor
string (enum)
tile_resample

Options:

tile_resample, tile_colorfix+sharp, tile_colorfix, Без препроцессора

[ControlNet — tile/blur] Preprocessor
tile_blur_model
string (enum)
controlnet11Models_tile

Options:

controlnet11Models_tile, Без модели

[ControlNet — tile/blur] Model
tile_blur_control_weight
number
1

Max: 2

[ControlNet — tile/blur] Control Weight
tile_blur_guidance_start
number
0

Max: 1

[ControlNet — tile/blur] Starting Control Step
tile_blur_guidance_end
number
1

Max: 1

[ControlNet — tile/blur] Ending Control Step
tile_blur_down_sampling_rate
number
1

Min: 1

Max: 8

[ControlNet — tile/blur] Down Sampling Rate
instruct_p2p_enabled
boolean
False
[ControlNet — InstructP2P] Enable
instruct_p2p_image
string
[ControlNet — InstructP2P] Input image
instruct_p2p_use_image_size
boolean
False
[ControlNet — InstructP2P] Use provided image size
instruct_p2p_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — InstructP2P] Control Mode
instruct_p2p_model
string (enum)
control_v11e_sd15_ip2p_fp16

Options:

control_v11e_sd15_ip2p_fp16, Без модели

[ControlNet — InstructP2P] Model
instruct_p2p_control_weight
number
1

Max: 2

[ControlNet — InstructP2P] Control Weight
instruct_p2p_guidance_start
number
0

Max: 1

[ControlNet — InstructP2P] Starting Control Step
instruct_p2p_guidance_end
number
1

Max: 1

[ControlNet — InstructP2P] Ending Control Step
reference_enabled
boolean
False
[ControlNet — reference] Enable
reference_image
string
[ControlNet — reference] Input image
reference_use_image_size
boolean
False
[ControlNet — reference] Use provided image size
reference_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — reference] Control Mode
reference_preprocessor
string (enum)
reference_only

Options:

reference_only, reference_adain, reference_adain+attn, Без препроцессора

[ControlNet — reference] Preprocessor
reference_control_weight
number
1

Max: 2

[ControlNet — reference] Control Weight
reference_guidance_start
number
0

Max: 1

[ControlNet — reference] Starting Control Step
reference_guidance_end
number
1

Max: 1

[ControlNet — reference] Ending Control Step
reference_style_fidelity
number
0.5

Max: 1

[ControlNet — reference] Style Fidelity [only "Balanced" control mode]
reference_2_enabled
boolean
False
[ControlNet — reference — 2] Enable
reference_2_image
string
[ControlNet — reference — 2] Input image
reference_2_use_image_size
boolean
False
[ControlNet — reference — 2] Use provided image size
reference_2_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[ControlNet — reference — 2] Control Mode
reference_2_preprocessor
string (enum)
reference_only

Options:

reference_only, reference_adain, reference_adain+attn, Без препроцессора

[ControlNet — reference — 2] Preprocessor
reference_2_control_weight
number
1

Max: 2

[ControlNet — reference — 2] Control Weight
reference_2_guidance_start
number
0

Max: 1

[ControlNet — reference — 2] Starting Control Step
reference_2_guidance_end
number
1

Max: 1

[ControlNet — reference — 2] Ending Control Step
reference_2_style_fidelity
number
0.5

Max: 1

[ControlNet — reference — 2] Style Fidelity [only "Balanced" control mode]
lineart_enabled
boolean
False
[LineArt — openpose] Enable
lineart_image
string
[LineArt — openpose] Input image
lineart_use_image_size
boolean
False
[LineArt — openpose] Use provided image size
lineart_control_mode
string (enum)
Balanced

Options:

Balanced, My prompt is more important, ControlNet is more important

[LineArt — openpose] Control Mode
lineart_preprocessor
string (enum)
lineart_realistic

Options:

lineart_realistic, Без препроцессора

[LineArt — openpose] Preprocessor
lineart_model
string (enum)
controlnet11Models_lineart

Options:

controlnet11Models_lineart, Без модели

[LineArt — openpose] Model
lineart_control_weight
number
1

Max: 2

[LineArt — openpose] Control Weight
lineart_guidance_start
number
0

Max: 1

[LineArt — openpose] Starting Control Step
lineart_guidance_end
number
1

Max: 1

[LineArt — openpose] Ending Control Step
lineart_preprocessor_resolution
integer
512

Min: 64

Max: 2048

[LineArt — openpose] Preprocessor Resolution

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema 
{
  "type": "object",
  "title": "Output",
  "required": [
    "images",
    "seeds",
    "prompt"
  ],
  "properties": {
    "seeds": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "title": "Seeds"
    },
    "images": {
      "type": "array",
      "items": {
        "type": "string",
        "format": "uri"
      },
      "title": "Images"
    },
    "prompt": {
      "type": "string",
      "title": "Prompt"
    }
  }
}