Recraft Images API

REST API for AI image and vector generation. Endpoints cover prompt-to-image (raster and vector), image-to-image, inpainting, outpainting, background removal/replace/generate, vectorization, crisp and creative upscaling, region erase, image variation, exploration, prompt enhancement, custom style creation, and account/credit inspection. Authenticated with a Bearer API token from the Recraft profile; metered in API units ($1 = 1,000 units) and rate-limited at 100 images/min and 5 requests/sec per user.

Recraft Images API is one of 2 APIs that Recraft publishes on the APIs.io network, described by a machine-readable OpenAPI specification.

This API exposes 7 machine-runnable capabilities that can be deployed as REST, MCP, or Agent Skill surfaces via Naftiko and 2 JSON Schema definitions.

Tagged areas include AI, Artificial Intelligence, Image Generation, Vector Graphics, and Generative AI. The published artifact set on APIs.io includes API documentation, an OpenAPI specification, a JSON-LD context, 7 Naftiko capability specs, and 2 JSON Schemas.

Documentation

Specifications

Schemas & Data

Other Resources

OpenAPI Specification

recraft-images-api-openapi.yml Raw ↑
openapi: 3.0.3
info:
  title: Recraft Images API
  description: |
    Recraft's external HTTP API for AI image and vector generation, editing,
    transformation, vectorization, upscaling, background tooling, prompt
    enhancement, custom style creation, and account/credit inspection. All
    requests are authenticated with a Bearer API token issued from your
    Recraft profile.
  version: '2026-05-22'
  contact:
    name: Recraft
    url: https://www.recraft.ai
servers:
  - url: https://external.api.recraft.ai/v1
    description: Production
security:
  - bearerAuth: []
paths:
  /images/generations:
    post:
      operationId: createImage
      summary: Create Image From Prompt
      description: |
        Generate one or more images from a text prompt. Supports both raster
        and vector output across all Recraft model versions (V2, V3, V4, V4.1)
        and OpenAI-style request shape. Optionally accepts a `style`, `substyle`,
        `style_id` (for custom styles), `size`, and `controls` block with brand
        controls such as colors and background color.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/GenerationRequest'
      responses:
        '200':
          description: Image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/generations/raster:
    post:
      operationId: createRasterImage
      summary: Create Raster Image
      description: Generate raster-only images (PNG/WEBP/JPG). Restricts model selection to raster variants.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/GenerationRequest'
      responses:
        '200':
          description: Raster image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/generations/vector:
    post:
      operationId: createVectorImage
      summary: Create Vector Image
      description: Generate vector-only SVG images. Restricts model selection to vector variants.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/GenerationRequest'
      responses:
        '200':
          description: Vector image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/imageToImage:
    post:
      operationId: imageToImage
      summary: Transform Image To Image
      description: Generate a variation of an input image preserving composition while applying a new prompt or style.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/ImageToImageRequest'
      responses:
        '200':
          description: Transformed image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/inpaint:
    post:
      operationId: inpaintImage
      summary: Inpaint Masked Region
      description: Replace masked regions of an input image with content matching a prompt.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/MaskedEditRequest'
      responses:
        '200':
          description: Inpainted image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/outpaint:
    post:
      operationId: outpaintImage
      summary: Outpaint Image Beyond Edges
      description: Extend an image beyond its original boundaries using AI-generated content.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/OutpaintRequest'
      responses:
        '200':
          description: Outpainted image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/replaceBackground:
    post:
      operationId: replaceBackground
      summary: Replace Background
      description: Replace the detected background of an input image while preserving the subject.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/BackgroundRequest'
      responses:
        '200':
          description: Image set with replaced background
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/generateBackground:
    post:
      operationId: generateBackground
      summary: Generate Background Behind Subject
      description: Fill the masked background region with prompt-controlled generated content.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/MaskedEditRequest'
      responses:
        '200':
          description: Image set with generated background
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/vectorize:
    post:
      operationId: vectorizeImage
      summary: Vectorize Raster Image
      description: Convert a raster image to SVG vector format.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Vector image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/removeBackground:
    post:
      operationId: removeBackground
      summary: Remove Image Background
      description: Strip the background from an input image and return a transparent PNG.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Image set with background removed
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/crispUpscale:
    post:
      operationId: crispUpscale
      summary: Crisp Upscale Image
      description: Upscale an image with emphasis on sharpness and edge clarity.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Upscaled image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/creativeUpscale:
    post:
      operationId: creativeUpscale
      summary: Creative Upscale Image
      description: Upscale an image with emphasis on adding creative detail to faces and objects.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Upscaled image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/eraseRegion:
    post:
      operationId: eraseRegion
      summary: Erase Masked Region
      description: Remove the masked region from an image cleanly (object removal).
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/MaskedEditRequest'
      responses:
        '200':
          description: Image set with region erased
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/variateImage:
    post:
      operationId: variateImage
      summary: Variate Image
      description: Produce visual remixes of an input image preserving its overall feel.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Image variations
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/explore:
    post:
      operationId: exploreImages
      summary: Explore Diverse Variations
      description: Generate a wide range of visual variations from a prompt to support creative discovery.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/GenerationRequest'
      responses:
        '200':
          description: Image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /images/explore/similar:
    post:
      operationId: exploreSimilarImages
      summary: Explore Similar Images
      description: Generate images that resemble a source image while exploring stylistic neighbours.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/FileOnlyRequest'
      responses:
        '200':
          description: Image set
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageSet'
  /prompts/enhance:
    post:
      operationId: enhancePrompt
      summary: Enhance Prompt
      description: Expand a short user prompt with rich visual context to improve generation quality.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
              required: [prompt]
              properties:
                prompt:
                  type: string
                  description: Original prompt to enhance.
                model:
                  type: string
                  description: Optional target model the enhanced prompt will be used with.
      responses:
        '200':
          description: Enhanced prompt
          content:
            application/json:
              schema:
                type: object
                properties:
                  prompt:
                    type: string
  /styles:
    post:
      operationId: createStyle
      summary: Create Custom Style From Reference Images
      description: Upload up to five reference images to create a reusable custom style for V3 models. Returns a `style_id` to use in subsequent generation calls.
      requestBody:
        required: true
        content:
          multipart/form-data:
            schema:
              $ref: '#/components/schemas/CreateStyleRequest'
      responses:
        '200':
          description: Style reference
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Style'
  /users/me:
    get:
      operationId: getCurrentUser
      summary: Get Current User And Credit Balance
      description: Retrieve the authenticated user's profile, plan, and remaining API unit balance.
      responses:
        '200':
          description: Current user
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/User'
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: Recraft API token issued from the Recraft profile. Sent as `Authorization`.
  schemas:
    GenerationRequest:
      type: object
      required: [prompt]
      properties:
        prompt:
          type: string
          maxLength: 10000
          description: Text prompt describing the desired image (up to 10,000 chars for V4/V4.1; 1,000 for V2/V3).
        model:
          type: string
          enum:
            - recraftv4_1
            - recraftv4_1_pro
            - recraftv4_1_utility
            - recraftv4_1_utility_pro
            - recraftv4
            - recraftv4_pro
            - recraftv3
            - recraftv2
          default: recraftv4_1
          description: Recraft model identifier.
        style:
          type: string
          description: Style group such as `realistic_image`, `digital_illustration`, `vector_illustration`, or `logo_raster`.
        substyle:
          type: string
          description: Optional specific substyle within the selected style (e.g. `pixel_art`, `linocut`, `urban_drama`).
        style_id:
          type: string
          format: uuid
          description: Identifier of a custom style created via `/styles`.
        size:
          type: string
          description: Output size; either an aspect ratio (e.g. `1024x1024`) or one of 14 supported ratios (1:1, 2:1, 1:2, 3:2, 2:3, 4:3, 3:4, 5:4, 4:5, 6:10, 14:10, 10:14, 16:9, 9:16).
        n:
          type: integer
          minimum: 1
          maximum: 6
          default: 1
        response_format:
          type: string
          enum: [url, b64_json]
          default: url
        negative_prompt:
          type: string
        controls:
          $ref: '#/components/schemas/BrandControls'
        text_layout:
          type: array
          description: Optional positional control for text on an image.
          items:
            type: object
            properties:
              text:
                type: string
              bbox:
                type: array
                items:
                  type: number
    BrandControls:
      type: object
      description: Brand-level controls applied during generation.
      properties:
        colors:
          type: array
          items:
            $ref: '#/components/schemas/Color'
        background_color:
          $ref: '#/components/schemas/Color'
        no_text:
          type: boolean
        artistic_level:
          type: integer
          minimum: 0
          maximum: 5
    Color:
      type: object
      required: [rgb]
      properties:
        rgb:
          type: array
          minItems: 3
          maxItems: 3
          items:
            type: integer
            minimum: 0
            maximum: 255
        std:
          type: number
          minimum: 0
          maximum: 1
    ImageToImageRequest:
      type: object
      required: [image, prompt]
      properties:
        image:
          type: string
          format: binary
        prompt:
          type: string
        strength:
          type: number
          minimum: 0
          maximum: 1
          default: 0.5
        model:
          type: string
        style:
          type: string
        substyle:
          type: string
        style_id:
          type: string
          format: uuid
        n:
          type: integer
        response_format:
          type: string
          enum: [url, b64_json]
        negative_prompt:
          type: string
    MaskedEditRequest:
      allOf:
        - $ref: '#/components/schemas/ImageToImageRequest'
        - type: object
          required: [mask]
          properties:
            mask:
              type: string
              format: binary
              description: Binary mask image. White pixels mark the region to edit.
    OutpaintRequest:
      type: object
      required: [image]
      properties:
        image:
          type: string
          format: binary
        prompt:
          type: string
        outpaint_top:
          type: integer
        outpaint_bottom:
          type: integer
        outpaint_left:
          type: integer
        outpaint_right:
          type: integer
        model:
          type: string
        style:
          type: string
    BackgroundRequest:
      type: object
      required: [image]
      properties:
        image:
          type: string
          format: binary
        prompt:
          type: string
        model:
          type: string
        style:
          type: string
        n:
          type: integer
        response_format:
          type: string
          enum: [url, b64_json]
    FileOnlyRequest:
      type: object
      required: [image]
      properties:
        image:
          type: string
          format: binary
        response_format:
          type: string
          enum: [url, b64_json]
    CreateStyleRequest:
      type: object
      required: [style, file]
      properties:
        style:
          type: string
          description: Base style group (e.g. `digital_illustration`).
        file:
          type: array
          minItems: 1
          maxItems: 5
          items:
            type: string
            format: binary
    Style:
      type: object
      properties:
        id:
          type: string
          format: uuid
        credits:
          type: integer
    ImageSet:
      type: object
      properties:
        created:
          type: integer
          format: int64
        credits:
          type: integer
        data:
          type: array
          items:
            $ref: '#/components/schemas/Image'
    Image:
      type: object
      properties:
        image_id:
          type: string
          format: uuid
        url:
          type: string
          format: uri
          description: Signed URL valid for approximately 24 hours.
        b64_json:
          type: string
        revised_prompt:
          type: string
    User:
      type: object
      properties:
        id:
          type: string
        email:
          type: string
          format: email
        credits:
          type: integer
          description: Remaining API unit balance (1,000 units = USD $1.00).