Mathpix

Mathpix Document OCR API

Asynchronous PDF and document OCR via v3/pdf. Submit by URL or multipart upload, poll status, optionally stream via SSE, and download outputs in MMD, Markdown, DOCX, LaTeX (.tex.zip), HTML, PPTX, or per-line JSON. Supports page ranges, alphabets, chemistry (SMILES), idiomatic equation arrays, and page-info capture.

Mathpix Document OCR API is one of 7 APIs that Mathpix publishes on the APIs.io network, described by a machine-readable OpenAPI specification.

This API exposes 1 machine-runnable capability that can be deployed as REST, MCP, or Agent Skill surfaces via Naftiko and 1 JSON Schema definition.

Tagged areas include OCR, PDF, Documents, STEM, and Conversion. The published artifact set on APIs.io includes API documentation, an OpenAPI specification, sample payloads, 1 Naftiko capability spec, and 1 JSON Schema.

GitHub OpenAPI

Documentation

📖
Documentation
https://docs.mathpix.com/reference/post-v3-pdf

Specifications

⚙
OpenAPI
https://raw.githubusercontent.com/api-evangelist/mathpix/refs/heads/main/openapi/mathpix-document-ocr-api-openapi.yml

Examples

📝
Example
https://raw.githubusercontent.com/api-evangelist/mathpix/refs/heads/main/examples/mathpix-process-document-example.json

Schemas & Data

📊
JSONSchema
https://raw.githubusercontent.com/api-evangelist/mathpix/refs/heads/main/json-schema/mathpix-document-result-schema.json

Other Resources

🔗
NaftikoCapability
https://raw.githubusercontent.com/api-evangelist/mathpix/refs/heads/main/capabilities/document-ocr-documents.yaml

OpenAPI Specification

mathpix-document-ocr-api-openapi.yml Raw ↑ 
openapi: 3.1.0
info:
  title: Mathpix Document OCR API
  description: >
    Mathpix v3/pdf endpoints for asynchronous document OCR and format conversion. Submit PDFs
    (and other supported documents) by URL or multipart upload, poll status, and download results
    in MMD, Markdown, DOCX, LaTeX (.tex.zip), HTML, PPTX, or line-by-line JSON. Includes server-sent
    event streaming for incremental results.
  version: v3
  contact:
    name: Mathpix Support
    url: https://docs.mathpix.com
    email: [email protected]
  license:
    name: Mathpix Terms of Service
    url: https://mathpix.com/terms-of-service
servers:
  - url: https://api.mathpix.com
    description: Production Server
security:
  - AppIdAuth: []
    AppKeyAuth: []
tags:
  - name: Documents
    description: Submit, monitor, and retrieve document OCR jobs.
paths:
  /v3/pdf:
    post:
      summary: Process A Document
      description: >
        Submit a document for asynchronous OCR. Accepts a public URL via JSON body or a multipart
        file upload. Returns a pdf_id used to poll status and download results.
      operationId: processDocument
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/PdfRequest'
          multipart/form-data:
            schema:
              type: object
              properties:
                file:
                  type: string
                  format: binary
                  description: The document binary to OCR.
                options_json:
                  type: string
                  description: JSON-encoded request options matching PdfRequest fields.
      responses:
        '200':
          description: Job accepted.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/PdfSubmitResponse'
        '401':
          description: Missing or invalid credentials.
  /v3/pdf/{pdf_id}:
    get:
      summary: Get Document Status
      description: Check processing status for a previously submitted document.
      operationId: getDocumentStatus
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Job status.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/PdfStatus'
    delete:
      summary: Delete Document
      description: Permanently delete the submitted document and its derived outputs.
      operationId: deleteDocument
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Deletion confirmed.
  /v3/pdf/{pdf_id}/stream:
    get:
      summary: Stream Document Results
      description: Stream OCR results incrementally via server-sent events as pages complete.
      operationId: streamDocument
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: text/event-stream of incremental results.
          content:
            text/event-stream:
              schema:
                type: string
  /v3/pdf/{pdf_id}.mmd:
    get:
      summary: Download Mathpix Markdown
      operationId: downloadMmd
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Mathpix Markdown file.
          content:
            text/plain:
              schema:
                type: string
  /v3/pdf/{pdf_id}.md:
    get:
      summary: Download Standard Markdown
      operationId: downloadMd
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Standard Markdown file.
          content:
            text/plain:
              schema:
                type: string
  /v3/pdf/{pdf_id}.docx:
    get:
      summary: Download DOCX
      operationId: downloadDocx
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Microsoft Word document.
          content:
            application/vnd.openxmlformats-officedocument.wordprocessingml.document:
              schema:
                type: string
                format: binary
  /v3/pdf/{pdf_id}.tex.zip:
    get:
      summary: Download LaTeX Archive
      operationId: downloadTexZip
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Zipped LaTeX source.
          content:
            application/zip:
              schema:
                type: string
                format: binary
  /v3/pdf/{pdf_id}.html:
    get:
      summary: Download HTML
      operationId: downloadHtml
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: HTML document.
          content:
            text/html:
              schema:
                type: string
  /v3/pdf/{pdf_id}.pptx:
    get:
      summary: Download PPTX
      operationId: downloadPptx
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: PowerPoint document.
          content:
            application/vnd.openxmlformats-officedocument.presentationml.presentation:
              schema:
                type: string
                format: binary
  /v3/pdf/{pdf_id}.lines.json:
    get:
      summary: Download Lines JSON
      operationId: downloadLinesJson
      tags:
        - Documents
      parameters:
        - $ref: '#/components/parameters/AppIdHeader'
        - $ref: '#/components/parameters/AppKeyHeader'
        - $ref: '#/components/parameters/PdfIdPath'
      responses:
        '200':
          description: Line-by-line JSON data.
          content:
            application/json:
              schema:
                type: object
                additionalProperties: true
components:
  securitySchemes:
    AppIdAuth:
      type: apiKey
      in: header
      name: app_id
    AppKeyAuth:
      type: apiKey
      in: header
      name: app_key
  parameters:
    AppIdHeader:
      name: app_id
      in: header
      required: true
      schema:
        type: string
    AppKeyHeader:
      name: app_key
      in: header
      required: true
      schema:
        type: string
    PdfIdPath:
      name: pdf_id
      in: path
      required: true
      schema:
        type: string
      description: Job identifier returned by POST /v3/pdf.
  schemas:
    PdfRequest:
      type: object
      properties:
        url:
          type: string
          format: uri
          description: HTTPS URL to download the document from. Mutually exclusive with multipart file upload.
        streaming:
          type: boolean
          description: When true, results are emitted via SSE on /v3/pdf/{pdf_id}/stream as pages complete.
          default: false
        metadata:
          type: object
          additionalProperties: true
        alphabets_allowed:
          type: object
          description: Restrict the OCR output to specific alphabets (e.g. {"en": true, "ru": false}).
          additionalProperties:
            type: boolean
        rm_spaces:
          type: boolean
          default: true
        rm_fonts:
          type: boolean
          default: false
        idiomatic_eqn_arrays:
          type: boolean
          default: false
        include_equation_tags:
          type: boolean
        include_smiles:
          type: boolean
          default: true
        include_page_breaks:
          type: boolean
          default: false
        include_page_info:
          type: boolean
          default: false
        page_ranges:
          type: string
          description: Comma-separated page selection (e.g. "2,4-6").
        conversion_formats:
          type: object
          description: Output formats to generate, e.g. {"docx": true, "tex.zip": true, "md": true, "html": true, "pptx": true}.
          additionalProperties:
            type: boolean
        conversion_options:
          type: object
          additionalProperties: true
    PdfSubmitResponse:
      type: object
      properties:
        pdf_id:
          type: string
          description: Identifier used to poll status and download outputs.
    PdfStatus:
      type: object
      properties:
        status:
          type: string
          enum: [received, loaded, split, completed, processing, error]
        num_pages:
          type: integer
        percent_done:
          type: number
          format: float
        error:
          type: string
        conversion_status:
          type: object
          description: Per-format conversion progress map (e.g. {"docx": {"status": "completed"}}).
          additionalProperties:
            type: object
            additionalProperties: true