import requests

response = requests.post(
    f"https://api.stability.ai/v2alpha/generation/stable-image/inpaint",
    headers={
        "authorization": f"Bearer sk-MYAPIKEY"
    },
    files={
        "image": open("./husky-in-a-field.png", "rb")
    },
    data={
        "prompt": "golden retriever in a field",
        "mode": "search",
        "search_prompt": "dog",
        "output_format": "webp",
    },
)

if response.status_code == 200:
    with open("./golden-retriever-in-a-field.webp", 'wb') as file:
        file.write(response.content)
else:
    raise Exception(str(response.json()))

"The bytes of the generated jpeg"

{
  "id": "a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4",
  "name": "bad_request",
  "errors": [
    "some-field: is required"
  ]
}

{
  "id": "ed14db44362126aab3cbd25cca51ffe3",
  "name": "content_moderation",
  "errors": [
    "Your request was flagged by our content moderation system, as a result your request was denied and you were not charged."
  ]
}

{
  "id": "2a1b2d4eafe2bc6ab4cd4d5c6133f513",
  "name": "internal_error",
  "errors": [
    "An unexpected server error has occurred, please try again later."
  ]
}

v2alpha/generation

stable-image/inpaint

Inpaint an existing image, with or without a mask, using our latest-and-greatest inpainting model.

Search-and-Replace Mode

This mode is ideal for individuals of all levels of skill in design. It can be used for straightforward adjustments to images. The service will automatically mask the most appropriate object based on the contents of the search_prompt, and replace it with a generated result based on the prompt.

How to use: set the mode parameter to search and provide a short description of what to search-and-replace in the search_prompt parameter.

Mask Mode

This mode allows for precise control of generative fill tasks on an image, down to the level of individual pixels. Design professionals can provide a mask for the section of the image to be replaced, and use standard image prompting to describe the full image as it should appear after the editing. The resulting image will incorporate all of the elements described in the prompt.

How to use: set the mode parameter to mask and either pass in an image with an alpha channel or provide an explicit mask image to the mask parameter. If both are present the mask parameter will take precedence.

Price

Requests with mode set to search cost 4 cents.
Requests with mode set to mask cost 3 cents.

POST

v2alpha

generation

stable-image

inpaint

import requests

response = requests.post(
    f"https://api.stability.ai/v2alpha/generation/stable-image/inpaint",
    headers={
        "authorization": f"Bearer sk-MYAPIKEY"
    },
    files={
        "image": open("./husky-in-a-field.png", "rb")
    },
    data={
        "prompt": "golden retriever in a field",
        "mode": "search",
        "search_prompt": "dog",
        "output_format": "webp",
    },
)

if response.status_code == 200:
    with open("./golden-retriever-in-a-field.webp", 'wb') as file:
        file.write(response.content)
else:
    raise Exception(str(response.json()))

"The bytes of the generated jpeg"

{
  "id": "a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4",
  "name": "bad_request",
  "errors": [
    "some-field: is required"
  ]
}

{
  "id": "ed14db44362126aab3cbd25cca51ffe3",
  "name": "content_moderation",
  "errors": [
    "Your request was flagged by our content moderation system, as a result your request was denied and you were not charged."
  ]
}

{
  "id": "2a1b2d4eafe2bc6ab4cd4d5c6133f513",
  "name": "internal_error",
  "errors": [
    "An unexpected server error has occurred, please try again later."
  ]
}

Authorizations

authorization

string

header

required

Use your Stability API key to authentication requests to this App.

Headers

authorization

string

required

Your Stability API key, used to authenticate your requests. Although you may have multiple keys in your account, you should use the same key for all requests to this API.

Minimum string length: 1

content-type

string

required

The content type of the request body. Do not manually specify this header; your HTTP client library will automatically include the appropriate boundary parameter.

Minimum string length: 1

Example:

"multipart/form-data"

enum<string>

default:image/*

Specify image/* to get the image bytes directly. Otherwise specify application/json to receive the image as base64 encoded JSON.

Available options:

image/*,

application/json

Body

multipart/form-data

Option 1
Option 2

mode

enum<string>

required

Controls how the model decides which areas to inpaint and which areas to leave alone.

Specifying mask requires:

Provide an explicit mask image in the mask parameter
Use the alpha channel of the image parameter as the mask

Specifying search requires:

Provide a small description of what to inpaint in the search_prompt parameter

Available options:

search

search_prompt

string

required

Short description of what to inpaint in the image.

Example:

"glasses"

image

file

required

The image you wish to inpaint.

Supported Formats:

jpeg
png
webp

Validation Rules:

Every side must be at least 64 pixels
Total pixel count must be between 4,096 and 9,437,184 pixels

prompt

string

required

What you wish to see in the output image. A strong, descriptive prompt that clearly defines elements, colors, and subjects will lead to better results.

To control the weight of a given word use the format (word:weight), where word is the word you'd like to control the weight of and weight is a value between 0 and 1. For example: The sky was a crisp (blue:0.3) and (green:0.8) would convey a sky that was blue and green, but more green than blue.

Required string length: 1 - 10000

negative_prompt

string

A blurb of text describing what you do not wish to see in the output image. This is an advanced feature.

Maximum string length: 10000

seed

number

default:0

A specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass 0 to use a random seed.)

Required range: 0 <= x <= 4294967294

output_format

enum<string>

default:png

Dictates the content-type of the generated image.

Available options:

jpeg,

png,

webp

Response

Inpainting was successful.

The bytes of the generated image.

The finish-reason and seed will be present as headers.