Mistral AI OCR API

openapi: 3.1.0
info:
  title: Mistral AI OCR API
  description: >-
    The Mistral AI OCR API provides optical character recognition capabilities
    powered by the mistral-ocr-latest model. It can extract text and structured
    content from PDF documents and images, comprehending complex document
    elements including media, tables, equations, and interleaved text. The API
    returns ordered, structured content suitable for downstream processing and
    retrieval augmented generation pipelines.
  version: '1.0.0'
  contact:
    name: Mistral AI Support
    url: https://docs.mistral.ai
  termsOfService: https://mistral.ai/terms
externalDocs:
  description: Mistral AI OCR Documentation
  url: https://docs.mistral.ai/api/endpoint/ocr
servers:
  - url: https://api.mistral.ai/v1
    description: Mistral AI Production Server
tags:
  - name: OCR
    description: >-
      Endpoints for extracting text and structured content from documents
      and images using optical character recognition.
security:
  - bearerAuth: []
paths:
  /ocr:
    post:
      operationId: processOcr
      summary: Process document with OCR
      description: >-
        Extracts text and structured content from a PDF document or image
        using Mistral's OCR model. The API comprehends complex document
        elements including tables, equations, charts, and interleaved text,
        returning structured markdown content organized by page.
      tags:
        - OCR
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/OcrRequest'
      responses:
        '200':
          description: Successful OCR processing response
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/OcrResponse'
        '400':
          description: Bad request due to invalid parameters
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '401':
          description: Unauthorized due to missing or invalid API key
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
        '429':
          description: Rate limit exceeded
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/Error'
components:
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      bearerFormat: API Key
  schemas:
    OcrRequest:
      type: object
      required:
        - model
        - document
      properties:
        model:
          type: string
          description: >-
            ID of the model to use for OCR processing.
          example: mistral-ocr-latest
        document:
          $ref: '#/components/schemas/DocumentInput'
        pages:
          type: array
          description: >-
            Specific page numbers to process. If not provided, all pages
            are processed.
          items:
            type: integer
            minimum: 0
        include_image_base64:
          type: boolean
          description: >-
            Whether to include base64-encoded images of extracted figures
            and charts in the response.
          default: false
        image_limit:
          type: integer
          description: >-
            Maximum number of images to extract and return.
          minimum: 0
        image_min_size:
          type: integer
          description: >-
            Minimum size in pixels for images to be included in the response.
          minimum: 0
    DocumentInput:
      type: object
      required:
        - type
      properties:
        type:
          type: string
          description: >-
            The type of document input.
          enum:
            - document_url
            - image_url
            - base64
        document_url:
          type: string
          format: uri
          description: >-
            URL of the document to process. Used when type is document_url.
        image_url:
          type: string
          format: uri
          description: >-
            URL of the image to process. Used when type is image_url.
        content:
          type: string
          description: >-
            Base64-encoded content of the document. Used when type is base64.
    OcrResponse:
      type: object
      properties:
        pages:
          type: array
          description: >-
            A list of pages with extracted content.
          items:
            $ref: '#/components/schemas/OcrPage'
        model:
          type: string
          description: >-
            The model used for OCR processing.
        usage:
          $ref: '#/components/schemas/OcrUsage'
    OcrPage:
      type: object
      properties:
        index:
          type: integer
          description: >-
            The zero-based page index in the document.
        markdown:
          type: string
          description: >-
            The extracted content in markdown format, preserving document
            structure including headers, tables, and equations.
        images:
          type: array
          description: >-
            Extracted images from the page, if include_image_base64 was set.
          items:
            $ref: '#/components/schemas/ExtractedImage'
        dimensions:
          $ref: '#/components/schemas/PageDimensions'
    ExtractedImage:
      type: object
      properties:
        id:
          type: string
          description: >-
            A unique identifier for the extracted image.
        base64:
          type: string
          description: >-
            Base64-encoded image data.
        content_type:
          type: string
          description: >-
            The MIME type of the image.
    PageDimensions:
      type: object
      properties:
        width:
          type: integer
          description: >-
            Width of the page in pixels.
        height:
          type: integer
          description: >-
            Height of the page in pixels.
    OcrUsage:
      type: object
      properties:
        pages_processed:
          type: integer
          description: >-
            Number of pages processed.
        doc_size_bytes:
          type: integer
          description: >-
            Size of the document in bytes.
    Error:
      type: object
      properties:
        message:
          type: string
          description: >-
            A human-readable error message.
        type:
          type: string
          description: >-
            The type of error.
        code:
          type: integer
          description: >-
            The HTTP status code.