> ## Documentation Index > Fetch the complete documentation index at: https://docs.stability.ai/llms.txt > Use this file to discover all available pages before exploring further. # 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. ## OpenAPI ````yaml /openapi.json post /v2alpha/generation/stable-image/inpaint openapi: 3.0.1 info: version: v2beta title: StabilityAI REST API description: >- Welcome to the Stability Platform API. As of March 2024, we are building the REST v2beta API service to be the primary API service for the Stability Platform. All AI services on other APIs (gRPC, REST v1, RESTv2alpha) will continue to be maintained, however they will not receive new features or parameters. If you are a REST v2alpha user, we strongly recommend that you adjust the URL calls for the specific services that you are using over to the equivalent REST v2beta URL. Normally, this means simply replacing "v2alpha" with "v2beta". We are not deprecating v2alpha URLs at this time for users that are currently using them. #### Authentication You will need your [Stability API key](https://platform.stability.ai/account/keys) in order to make requests to this API. Make sure you never share your API key with anyone, and you never commit it to a public repository. Include this key in the `Authorization` header of your requests. #### Rate limiting This API is rate-limited to 150 requests every 10 seconds. If you exceed this limit, you will receive a `429` response and be timed out for 60 seconds. If you find this limit too restrictive, please reach out to us via [this form](https://kb.stability.ai/knowledge-base/kb-tickets/new). #### Support Please see our [FAQ](https://platform.stability.ai/faq) for answers to common questions. If you have any other questions or concerns, please reach out to us via [this form](https://kb.stability.ai/knowledge-base/kb-tickets/new). To see the health of our APIs, please check our [Status Page](https://stabilityai.instatus.com/). servers: - url: https://api.stability.ai security: - STABILITY_API_KEY: [] tags: - name: Edit description: >- Tools for editing your own and generated images. **[Erase](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1erase/post)** The Erase service removes unwanted objects, such as blemishes on portraits or items on desks, using image masks. **[Outpaint](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1outpaint/post)** The outpaint service inserts additional content in an image to fill in the space in any direction, allowing you to "zoom-out" of an image. **[Inpaint](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1inpaint/post)** The Inpaint service modifies images by filling in or replacing specified areas with new content based on the content of a "mask" image. **[Search and Replace](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1search-and-replace/post)** The Search and Replace service, similar to inpaint, allows to replace specified areas with new content, but this time with the help of a prompt instead of a mask. The service will automatically segment the object and replace it with the object requested in the prompt. **[Search and Recolor](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1search-and-recolor/post)** The Search and Recolor service is another derivative of the inpaint service and provides the ability to change the color of a specific object in an image using a prompt. The Search and Recolor service will automatically segment the object and recolor it using the colors requested in the prompt. **[Remove Background](/docs/api-reference#tag/Edit/paths/~1v2beta~1stable-image~1edit~1remove-background/post)** The Remove Background service accurately segments the foreground from an image to removes the background. - name: Upscale description: >- Tools for increasing the size and resolution of your existing images. **[Fast Upscaler](/docs/api-reference#tag/Upscale/paths/~1v2beta~1stable-image~1upscale~1fast/post)** This service enhances image resolution by 4x using predictive and generative AI. This lightweight and fast service (processing in ~1 second) is ideal for enhancing the quality of compressed images, making it suitable for social media posts and other applications. **[Conservative Upscaler](/docs/api-reference#tag/Upscale/paths/~1v2beta~1stable-image~1upscale~1conservative/post)** This service can upscale images by 20 to 40 times up to a 4 megapixel output image with minimal alteration to the original image. The Conservative Upscaler can upscale images as small as 64x64 pixels directly to a 4 megapixel output. Use this option if you directly need a 4 megapixel output. **[Creative Upscaler](/docs/api-reference#tag/Upscale/paths/~1v2beta~1stable-image~1upscale~1creative/post)** The service can upscale highly degraded images (lower than 1 megapixel) with a creative twist to provide high resolution results. - name: Generate description: >- Tools to generate new images from text, or create variations of existing images. Our different services include: **[Stable Image Ultra](/docs/api-reference#tag/Generate/paths/~1v2beta~1stable-image~1generate~1ultra/post)**: Photorealistic, Large-Scale Output Our state of the art text to image model based on Stable Diffusion 3.5. Stable Image Ultra Produces the highest quality, photorealistic outputs perfect for professional print media and large format applications. Stable Image Ultra excels at rendering exceptional detail and realism. **[Stable Image Core](/docs/api-reference#tag/Generate/paths/~1v2beta~1stable-image~1generate~1core/post)**: Fast and Affordable Optimized for fast and affordable image generation, great for rapidly iterating on concepts during ideation. Stable Image Core is the next generation model following Stable Diffusion XL. **[Stable Diffusion 3.5 Model Suite](/docs/api-reference#tag/Generate/paths/~1v2beta~1stable-image~1generate~1sd3/post)**: Stability AI's latest base models The different versions of our open models are available via API, letting you test and adjust speed and quality based on your use case. All model versions strike a balance between generation speed and output quality and are ideal for creating high-volume, high-quality digital assets like websites, newsletters, and marketing materials. - name: Control description: >- Tools for generating precise, controlled variations of existing images or sketches. **[Sketch](/docs/api-reference#tag/Control/paths/~1v2beta~1stable-image~1control~1sketch/post)** This service upgrades sketches to refined outputs with precise control. For non-sketch images, it allows detailed manipulation of the final appearance by leveraging the contour lines and edges within the image. **[Structure](/docs/api-reference#tag/Control/paths/~1v2beta~1stable-image~1control~1structure/post)** This service excels in generating images by maintaining the structure of an input image, making it especially valuable for advanced content creation scenarios such as recreating scenes or rendering characters from models. **[Style](/docs/api-reference#tag/Control/paths/~1v2beta~1stable-image~1control~1style/post)** This service extracts stylistic elements from an input image (control image) and uses it to guide the creation of an output image based on the prompt. The result is a new image in the same style as the control image. - name: Results description: Tools for fetching the results of your async generations. - name: Stable Audio 2 description: >- Tools to generate music and sound from text or audio, or transform existing audio clips into new compositions. Our different services include: **Stable Audio 2.5**: Fast, Best-Quality, Long-Form Music & Audio Generation Our most advanced audio generation model, capable of generating up to 3-minute, 44.1 kHz stereo compositions. Stable Audio 2.5 supports text-to-audio, audio-to-audio, and audio-inpaint workflows - allowing creators to upload a sound and transform it into new instruments, styles, or genres using natural language prompts. Ideal for music production, cinematic sound design, and remixing. **Stable Audio 2.0**: High-Quality Audio Generation Built for text-to-audio and audio-to-audio generation, also capable of generating up to 3-minute, 44.1 kHz stereo. Stable Audio 2.0 is great for ideation, music demos, and ambient soundscapes. It's optimized for creative professionals seeking detailed and extended outputs from simple prompts. Stable Audio models were exclusively trained on licensed data from the [AudioSparx](https://www.audiosparx.com/) music library, honoring opt-out requests and ensuring fair compensation for creators. Additionally, Stable Audio 2.5 was pre-trained on licensed data from [Freesound](https://freesound.org/). Read more about the model capabilities [here](https://stability.ai/news/stable-audio-2-0). x-displayName: Stable Audio 2.5 - name: Stable Audio description: >- Tools to generate music and sound from text or audio, or transform existing audio clips into new compositions. Our different services include: **Stable Audio 3.0**: Fast, Best-Quality, Long-Form Music & Audio Generation Our most advanced audio generation model, capable of generating up to 6-minute, 44.1 kHz stereo compositions. Stable Audio 3.0 supports text-to-audio, audio-to-audio, and audio-inpaint workflows - allowing creators to upload a sound and transform it into new instruments, styles, or genres using natural language prompts. Ideal for music production, cinematic sound design, and remixing. Stable Audio models were exclusively trained on licensed data from the [AudioSparx](https://www.audiosparx.com/) music library, honoring opt-out requests and ensuring fair compensation for creators. Additionally, Stable Audio 3.0 was pre-trained on licensed data from [Freesound](https://freesound.org/). Read more about the model capabilities [here](https://stability.ai/news/stable-audio-3-0). x-displayName: Stable Audio 3.0 paths: /v2alpha/generation/stable-image/inpaint: post: tags: - v2alpha/generation summary: stable-image/inpaint description: >- 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. parameters: - $ref: '#/components/parameters/Authorization' - $ref: '#/components/parameters/FormDataContentType' - schema: type: string default: image/* description: >- Specify `image/*` to get the image bytes directly. Otherwise specify `application/json` to receive the image as base64 encoded JSON. enum: - image/* - application/json required: false name: accept in: header requestBody: content: multipart/form-data: schema: oneOf: - $ref: '#/components/schemas/InpaintingSearchModeRequestBody' - $ref: '#/components/schemas/InpaintingMaskingModeRequestBody' discriminator: propertyName: mode mapping: search: '#/components/schemas/InpaintingSearchModeRequestBody' mask: '#/components/schemas/InpaintingMaskingModeRequestBody' responses: '200': $ref: '#/components/responses/ImageInpaintSuccess' '400': $ref: '#/components/responses/BadRequest' '403': $ref: '#/components/responses/ContentModerated' '500': $ref: '#/components/responses/InternalError' x-codeSamples: - lang: python label: Python source: |- import requests response = requests.post( f"https://api.stability.ai/v2alpha/generation/stable-image/inpaint", headers={ "authorization": f"Bearer sk-MYAPIKEY" }, files={ "image": open("./husky-in-a-field.png", "rb") }, data={ "prompt": "golden retriever in a field", "mode": "search", "search_prompt": "dog", "output_format": "webp", }, ) if response.status_code == 200: with open("./golden-retriever-in-a-field.webp", 'wb') as file: file.write(response.content) else: raise Exception(str(response.json())) - lang: javascript label: JavaScript source: |- import fs from "node:fs"; import axios from "axios"; import FormData from "form-data"; const formData = { image: fs.createReadStream("./husky-in-a-field.png"), prompt: "golden retriever standing in a field", mode: "search", search_prompt: "dog", output_format: "webp" }; const response = await axios.postForm( `https://api.stability.ai/v2alpha/generation/stable-image/inpaint`, axios.toFormData(formData, new FormData()), { validateStatus: undefined, responseType: "arraybuffer", headers: { Authorization: `Bearer sk-MYAPIKEY`, accept: "image/*" }, }, ); if(response.status === 200) { fs.writeFileSync("./golden-retriever-in-a-field.webp", Buffer.from(response.data)); } else { throw new Error(`${response.status}: ${response.data.toString()}`); } - lang: terminal label: cURL source: >- curl -f -sS "https://api.stability.ai/v2alpha/generation/stable-image/inpaint" \ -H "authorization: Bearer sk-MYAPIKEY" \ -H "accept: image/*" \ -F image=@"./husky-in-a-field.png" \ -F prompt="golden retriever in a field" \ -F mode="search" \ -F search_prompt="dog" \ -F output_format="webp" \ -o "./golden-retriever-in-a-field.webp" components: parameters: Authorization: schema: type: string description: >- Your [Stability API key](https://platform.stability.ai/account/keys), 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. minLength: 1 required: true name: authorization in: header FormDataContentType: schema: type: string minLength: 1 description: >- The content type of the request body. Do not manually specify this header; your HTTP client library will automatically include the appropriate boundary parameter. example: multipart/form-data required: true name: content-type in: header schemas: InpaintingSearchModeRequestBody: type: object properties: mode: type: string enum: - search description: >- 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 search_prompt: type: string description: Short description of what to inpaint in the `image`. example: glasses image: type: string description: |- 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 format: binary example: ./some/image.png prompt: $ref: '#/components/schemas/Prompt' negative_prompt: $ref: '#/components/schemas/NegativePrompt' seed: $ref: '#/components/schemas/Seed' output_format: type: string enum: - jpeg - png - webp default: png description: Dictates the `content-type` of the generated image. required: - image - prompt - mode - search_prompt InpaintingMaskingModeRequestBody: type: object properties: mode: type: string enum: - mask description: >- 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 mask: type: string description: >- Controls the strength of the inpainting process on a per-pixel basis, either via a second image (passed into this parameter) or via the alpha channel of the `image` parameter. **Passing in a Mask** The image passed to this parameter should be a black and white image that represents, at any pixel, the strength of inpainting based on how dark or light the given pixel is. Completely black pixels represent no inpainting strength while completely white pixels represent maximum strength. In the event the mask is a different size than the `image` parameter, it will be automatically resized. **Alpha Channel Support** If you don't provide an explicit mask, one will be derived from the alpha channel of the `image` parameter. Transparent pixels will be inpainted while opaque pixels will be preserved. In the event an `image` with an alpha channel is provided along with a `mask`, the `mask` will take precedence. format: binary example: ./some/image.png image: type: string description: |- 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 format: binary example: ./some/image.png prompt: $ref: '#/components/schemas/Prompt' negative_prompt: $ref: '#/components/schemas/NegativePrompt' seed: $ref: '#/components/schemas/Seed' output_format: type: string enum: - jpeg - png - webp default: png description: Dictates the `content-type` of the generated image. required: - image - prompt - mode Prompt: type: string minLength: 1 maxLength: 10000 description: >- 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. NegativePrompt: type: string maxLength: 10000 description: >- A blurb of text describing what you **do not** wish to see in the output image. This is an advanced feature. Seed: type: number minimum: 0 maximum: 4294967294 default: 0 description: >- A specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass `0` to use a random seed.) ImageBinaryResponse: type: string description: |- The bytes of the generated image. The `finish-reason` and `seed` will be present as headers. format: binary ImageJSONResponse: type: object properties: image: type: string description: The generated image, encoded to base64. example: AAAAIGZ0eXBpc29tAAACAGlzb21pc28yYXZjMW1... seed: allOf: - $ref: '#/components/schemas/Seed' - description: The seed used as random noise for this generation. example: 343940597 finish_reason: type: string enum: - SUCCESS - CONTENT_FILTERED description: >- The reason the generation finished. - `SUCCESS` = successful generation. - `CONTENT_FILTERED` = successful generation, however the output violated our content moderation policy and has been blurred as a result. example: SUCCESS required: - image - finish_reason Error: type: object properties: id: type: string minLength: 1 description: >- A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new) you file, as it will greatly assist us in diagnosing the root cause of the problem. example: a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4 name: type: string minLength: 1 description: >- Short-hand name for an error, useful for discriminating between errors with the same status code. example: bad_request errors: type: array items: type: string minItems: 1 description: One or more error messages indicating what went wrong. example: - 'some-field: is required' required: - id - name - errors ContentModerationResponse: type: object properties: id: type: string minLength: 1 description: >- A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new) you file, as it will greatly assist us in diagnosing the root cause of the problem. example: a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4 name: type: string minLength: 1 description: >- Our content moderation system has flagged some part of your request and subsequently denied it. You were not charged for this request. While this may at times be frustrating, it is necessary to maintain the integrity of our platform and ensure a safe experience for all users. If you would like to provide feedback, please use the [Support Form](https://kb.stability.ai/knowledge-base/kb-tickets/new). enum: - content_moderation errors: type: array items: type: string minItems: 1 description: One or more error messages indicating what went wrong. example: - 'some-field: is required' required: - id - name - errors description: Your request was flagged by our content moderation system. example: 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. responses: ImageInpaintSuccess: description: Inpainting was successful. headers: x-request-id: $ref: '#/components/headers/RequestID' content-type: description: |- The format of the generated image. To receive the bytes of the image directly, specify `image/*` in the accept header. To receive the bytes base64 encoded inside of a JSON payload, specify `application/json`. examples: jpeg: description: raw bytes value: image/jpeg jpegJSON: description: base64 encoded value: application/json; type=image/jpeg png: description: raw bytes value: image/png pngJSON: description: base64 encoded value: application/json; type=image/png webp: description: raw bytes value: image/webp webpJSON: description: base64 encoded value: application/json; type=image/webp schema: type: string finish-reason: $ref: '#/components/headers/ImageFinishReason' seed: $ref: '#/components/headers/Seed' content: image/jpeg: example: The bytes of the generated jpeg schema: $ref: '#/components/schemas/ImageBinaryResponse' application/json; type=image/jpeg: schema: $ref: '#/components/schemas/ImageJSONResponse' image/png: example: The bytes of the generated png schema: $ref: '#/components/schemas/ImageBinaryResponse' application/json; type=image/png: schema: $ref: '#/components/schemas/ImageJSONResponse' image/webp: example: The bytes of the generated webp schema: $ref: '#/components/schemas/ImageBinaryResponse' application/json; type=image/webp: schema: $ref: '#/components/schemas/ImageJSONResponse' BadRequest: description: Invalid parameter(s), see the `errors` field for details. content: application/json: schema: $ref: '#/components/schemas/Error' ContentModerated: description: Your request was flagged by our content moderation system. content: application/json: schema: $ref: '#/components/schemas/ContentModerationResponse' InternalError: description: >- An internal error occurred. If the problem persists [contact support](https://kb.stability.ai/knowledge-base/kb-tickets/new). content: application/json: schema: $ref: '#/components/schemas/Error' example: id: 2a1b2d4eafe2bc6ab4cd4d5c6133f513 name: internal_error errors: - An unexpected server error has occurred, please try again later. headers: RequestID: description: A unique identifier for this request. schema: type: string ImageFinishReason: schema: type: string enum: - SUCCESS - CONTENT_FILTERED description: >- Indicates the reason the generation finished. - `SUCCESS` = successful generation. - `CONTENT_FILTERED` = successful generation, however the output violated our content moderation policy and has been blurred as a result. > **NOTE:** This header is absent on JSON encoded responses because it is present in the body as `finish_reason`. Seed: description: >- The seed used as random noise for this generation. > **NOTE:** This header is absent on JSON encoded responses because it is present in the body as `seed`. example: '343940597' schema: type: string securitySchemes: STABILITY_API_KEY: type: apiKey name: authorization in: header description: >- Use your [Stability API key](https://platform.stability.ai/account/keys) to authentication requests to this App. ````