openapi: 3.1.0
info:
  title: NVIDIA NIM Image Generation API
  description: >
    Visual generative AI endpoints. Text-to-image, image-to-image, and image
    editing using models such as Black Forest Labs FLUX.1, Stable Diffusion XL,
    Shutterstock-trained Edify Image, and NVIDIA Edify 3D — exposed by per-model
    NIM containers with a uniform JSON contract that returns base64-encoded
    PNG or JPEG artifacts.
  version: '2026-05-25'
  contact:
    name: NVIDIA Developer Support
    url: https://forums.developer.nvidia.com/c/ai-data-science/nemo-llm-service/
  license:
    name: NVIDIA AI Enterprise License
    url: https://www.nvidia.com/en-us/data-center/products/ai-enterprise/
servers:
  - url: https://integrate.api.nvidia.com
    description: NVIDIA-hosted NIM endpoint
  - url: http://localhost:8000
    description: Self-hosted NIM container default
security:
  - BearerAuth: []
tags:
  - name: Images
    description: Text-to-image and image-to-image generation
paths:
  /v1/genai/{publisher}/{model}:
    post:
      summary: Generate An Image
      description: Generate or edit an image with a NIM-hosted visual generative model. The exact request schema varies by model family.
      operationId: generateImage
      tags:
        - Images
      parameters:
        - name: publisher
          in: path
          required: true
          schema:
            type: string
          description: Model publisher (e.g. `black-forest-labs`, `stabilityai`, `shutterstock`, `nvidia`).
        - name: model
          in: path
          required: true
          schema:
            type: string
          description: Model slug (e.g. `flux.1-schnell`, `sdxl-turbo`, `edify-image`).
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ImageGenerationRequest'
      responses:
        '200':
          description: Generated image artifacts.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ImageGenerationResponse'
        '400':
          description: Invalid request.
        '401':
          description: Missing or invalid API key.
        '429':
          description: Rate limit exceeded.
components:
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: nvapi-...
  schemas:
    ImageGenerationRequest:
      type: object
      required: [prompt]
      properties:
        prompt:
          type: string
        negative_prompt:
          type: string
        cfg_scale:
          type: number
          default: 5
        sampler:
          type: string
        seed:
          type: integer
        steps:
          type: integer
          default: 50
        width:
          type: integer
        height:
          type: integer
        mode:
          type: string
          enum: [text-to-image, image-to-image, inpaint]
          default: text-to-image
        image:
          type: string
          description: Optional base64-encoded source image for image-to-image / inpaint.
        mask:
          type: string
          description: Optional base64-encoded inpainting mask.
    ImageGenerationResponse:
      type: object
      properties:
        artifacts:
          type: array
          items:
            type: object
            properties:
              base64:
                type: string
                description: Base64-encoded image bytes (PNG or JPEG).
              finishReason:
                type: string
              seed:
                type: integer