Adobe Creative Suite Generative Fill API

openapi: 3.2.0
info:
  title: Adobe Creative Suite Adobe Firefly Generative Fill API
  description: The Adobe Firefly API provides access to Adobe's generative AI capabilities for creating and editing images, vectors, and video from natural language prompts. Built on Adobe's Firefly family of creative generative models, which are trained on licensed and public domain content. Supports text-to-image generation, generative fill, generative expand, generating similar images, object compositing, and video generation. All generation operations are asynchronous.
  version: 3.0.0
  termsOfService: https://www.adobe.com/legal/terms.html
  contact:
    name: Adobe Developer Support
    url: https://developer.adobe.com/support/
  license:
    name: Adobe Developer Terms
    url: https://www.adobe.com/legal/developer-terms.html
servers:
- url: https://firefly-api.adobe.io/v3
  description: Adobe Firefly API v3 production server
security:
- bearerAuth: []
tags:
- name: Generative Fill
  description: Fill selected regions of images with AI-generated content
paths:
  /images/fill-async:
    post:
      operationId: fillImageAsync
      summary: Adobe Creative Suite Generative Fill
      description: Fills a masked region of an existing image with AI-generated content based on a text prompt. A mask image defines the area to be filled. The generative fill respects the surrounding context to produce seamlessly blended results.
      tags:
      - Generative Fill
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ImageFillRequest'
      responses:
        '202':
          description: Fill job accepted and queued
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/AsyncJobSubmitted'
        '400':
          description: Bad request - invalid input parameters
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
        '401':
          description: Unauthorized - invalid or missing bearer token
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
      x-microcks-operation:
        delay: 0
        dispatcher: FALLBACK
components:
  schemas:
    InputImageReference:
      type: object
      description: Reference to an input image for generative operations
      required:
      - source
      properties:
        source:
          type: object
          description: Source of the input image
          properties:
            url:
              type: string
              description: Publicly accessible URL of the input image
              example: https://my-bucket.s3.amazonaws.com/source-image.jpg
            uploadId:
              type: string
              description: Upload ID for a previously uploaded image via the Firefly upload API
    ErrorResponse:
      type: object
      description: Standard error response body
      properties:
        code:
          type: string
          description: Machine-readable error code
          example: example_value
        message:
          type: string
          description: Human-readable description of the error
          example: example_value
        requestId:
          type: string
          description: Unique identifier for the failed request, for support purposes
          example: asset_abc123
    AsyncJobSubmitted:
      type: object
      description: Response returned immediately after an async generation job is submitted
      properties:
        jobId:
          type: string
          description: Unique identifier for the generation job
          example: urn:firefly:jobs:abc123def456
        statusUrl:
          type: string
          description: URL to poll for job status updates
          example: https://firefly-api.adobe.io/v3/status/urn:firefly:jobs:abc123def456
    ImageFillRequest:
      type: object
      description: Request body for generative fill operation
      required:
      - image
      - mask
      properties:
        prompt:
          type: string
          description: Text prompt describing what to fill in the masked area
          maxLength: 1024
          example: example_value
        negativePrompt:
          type: string
          description: Text describing what to avoid in the filled area
          maxLength: 1024
          example: example_value
        image:
          $ref: '#/components/schemas/InputImageReference'
        mask:
          $ref: '#/components/schemas/InputImageReference'
        numVariations:
          type: integer
          minimum: 1
          maximum: 4
          default: 1
          example: 1920
        seeds:
          type: array
          items:
            type: integer
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      bearerFormat: JWT
      description: OAuth 2.0 access token obtained from Adobe IMS