mlscade/stable_diffusion_inpaint_opt – Run with an API on Replicate

mlscade / stable_diffusion_inpaint_opt

stable_diffusion_inpaint_opt

Cold

Public
0 runs
A100 (80GB)

Run with an API

Playground API Examples README Versions

Input

generation_type

string

Generation mode

Default: "img2img"

clip_stop_at_last_layers

integer

(minimum: 1, maximum: 12)

Clip skip

Default: 2

upload_lora

string

Shift + Return to add a new line

Upload LoRA

prompt

string

Shift + Return to add a new line

Prompt

Default: ""

negative_prompt

string

Shift + Return to add a new line

(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(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

Default: "(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"

image

file

Input image

image_mask

file

Mask

mask_blur

integer

Mask blur

Default: 4

resize_mode

string

Resize mode

Default: "Just resize"

inpainting_mask_invert

string

Mask mode

Default: "Inpaint masked"

inpainting_fill

string

Masked content

Default: "Original"

inpaint_full_res

string

Inpaint area

Default: "Whole picture"

inpaint_full_res_padding

integer

Only masked padding, pixels

Default: 32

cfg_scale

number

CFG scale

Default: 7

denoising_strength

number

[HR] Denoising strength

Default: 0.75

seed

integer

Seed. Leave blank for random generation

width

integer

Width

Default: 512

height

integer

Height

Default: 512

num_outputs

integer

(minimum: 1, maximum: 4)

Batch size [Doesn't work in postprocess mode].

Default: 1

sampler

string

Sampling method

Default: "Euler a"

sampler_steps

integer

(minimum: 1, maximum: 50)

Sampling steps [Doesn't work in postprocess mode].

Default: 20

ad_enabled

boolean

[adetailer] Enable

Default: false

ad_model

string

[adetailer] ADetailer model

Default: "face_yolov8s"

ad_prompt

string

Shift + Return to add a new line

[adetailer] ADetailer prompt

Default: ""

ad_negative_prompt

string

Shift + Return to add a new line

[adetailer] ADetailer negative prompt

Default: ""

ad_use_sampler

boolean

[adetailer] Use separate sampler

Default: true

ad_sampler

string

[adetailer] ADetailer sampler

Default: "Euler a"

ad_confidence

number

[adetailer] Detection model confidence threshold

Default: 0.3

ad_dilate_erode

integer

[adetailer] Mask erosion (-) / dilation (+)

Default: 4

ad_mask_blur

integer

[adetailer] Inpaint mask blur

Default: 4

ad_mask_merge_mode

string

[adetailer] Mask merge mode

Default: "None"

ad_denoising_strength

number

[adetailer] Inpaint denoising strength

Default: 0.4

ad_inpaint_only_masked

boolean

[adetailer] Inpaint only masked

Default: true

ad_inpaint_only_masked_padding

integer

[adetailer] Inpaint only masked padding, pixels

Default: 32

ad_use_inpaint_width_height

boolean

[adetailer] Use separate width/height

Default: false

ad_inpaint_width

integer

[adetailer] Inpaint width

Default: 512

ad_inpaint_height

integer

[adetailer] Inpaint height

Default: 512

Run this model in Node.js with one line of code:

npx create-replicate --model=mlscade/stable_diffusion_inpaint_opt

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run mlscade/stable_diffusion_inpaint_opt using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "mlscade/stable_diffusion_inpaint_opt:af3d1f47be084111cf85fa89807c300272c832642ed6a5124c6bcef75eecba30",
  {
    input: {
      width: 512,
      height: 512,
      prompt: "",
      sampler: "Euler a",
      ad_model: "face_yolov8s",
      ad_prompt: "",
      cfg_scale: 7,
      mask_blur: 4,
      ad_enabled: false,
      ad_sampler: "Euler a",
      num_outputs: 1,
      resize_mode: "Just resize",
      ad_mask_blur: 4,
      ad_confidence: 0.3,
      sampler_steps: 20,
      ad_use_sampler: true,
      ad_dilate_erode: 4,
      generation_type: "img2img",
      inpainting_fill: "Original",
      negative_prompt: "(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",
      ad_inpaint_width: 512,
      inpaint_full_res: "Whole picture",
      ad_inpaint_height: 512,
      ad_mask_merge_mode: "None",
      ad_negative_prompt: "",
      denoising_strength: 0.75,
      ad_denoising_strength: 0.4,
      ad_inpaint_only_masked: true,
      inpainting_mask_invert: "Inpaint masked",
      clip_stop_at_last_layers: 2,
      inpaint_full_res_padding: 32,
      ad_use_inpaint_width_height: false,
      ad_inpaint_only_masked_padding: 32
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run mlscade/stable_diffusion_inpaint_opt using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "mlscade/stable_diffusion_inpaint_opt:af3d1f47be084111cf85fa89807c300272c832642ed6a5124c6bcef75eecba30",
    input={
        "width": 512,
        "height": 512,
        "prompt": "",
        "sampler": "Euler a",
        "ad_model": "face_yolov8s",
        "ad_prompt": "",
        "cfg_scale": 7,
        "mask_blur": 4,
        "ad_enabled": False,
        "ad_sampler": "Euler a",
        "num_outputs": 1,
        "resize_mode": "Just resize",
        "ad_mask_blur": 4,
        "ad_confidence": 0.3,
        "sampler_steps": 20,
        "ad_use_sampler": True,
        "ad_dilate_erode": 4,
        "generation_type": "img2img",
        "inpainting_fill": "Original",
        "negative_prompt": "(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",
        "ad_inpaint_width": 512,
        "inpaint_full_res": "Whole picture",
        "ad_inpaint_height": 512,
        "ad_mask_merge_mode": "None",
        "ad_negative_prompt": "",
        "denoising_strength": 0.75,
        "ad_denoising_strength": 0.4,
        "ad_inpaint_only_masked": True,
        "inpainting_mask_invert": "Inpaint masked",
        "clip_stop_at_last_layers": 2,
        "inpaint_full_res_padding": 32,
        "ad_use_inpaint_width_height": False,
        "ad_inpaint_only_masked_padding": 32
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run mlscade/stable_diffusion_inpaint_opt using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "mlscade/stable_diffusion_inpaint_opt:af3d1f47be084111cf85fa89807c300272c832642ed6a5124c6bcef75eecba30",
    "input": {
      "width": 512,
      "height": 512,
      "prompt": "",
      "sampler": "Euler a",
      "ad_model": "face_yolov8s",
      "ad_prompt": "",
      "cfg_scale": 7,
      "mask_blur": 4,
      "ad_enabled": false,
      "ad_sampler": "Euler a",
      "num_outputs": 1,
      "resize_mode": "Just resize",
      "ad_mask_blur": 4,
      "ad_confidence": 0.3,
      "sampler_steps": 20,
      "ad_use_sampler": true,
      "ad_dilate_erode": 4,
      "generation_type": "img2img",
      "inpainting_fill": "Original",
      "negative_prompt": "(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",
      "ad_inpaint_width": 512,
      "inpaint_full_res": "Whole picture",
      "ad_inpaint_height": 512,
      "ad_mask_merge_mode": "None",
      "ad_negative_prompt": "",
      "denoising_strength": 0.75,
      "ad_denoising_strength": 0.4,
      "ad_inpaint_only_masked": true,
      "inpainting_mask_invert": "Inpaint masked",
      "clip_stop_at_last_layers": 2,
      "inpaint_full_res_padding": 32,
      "ad_use_inpaint_width_height": false,
      "ad_inpaint_only_masked_padding": 32
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

No output yet! Press "Submit" to start a prediction.

Run time and cost

This model runs on Nvidia A100 (80GB) GPU hardware. We don't yet have enough runs of this model to provide performance information.

Readme

This model doesn't have a readme.