Jina AI Reader API

URL-to-markdown extraction

OpenAPI Specification

jina-ai-reader-api-openapi.yml Raw ↑
openapi: 3.0.3
info:
  title: Jina AI Embeddings Batch Reader API
  description: Generate high-quality multimodal embeddings for text, image, and code inputs using Jina AI's state-of-the-art embedding models. Supports synchronous embedding requests and batch jobs for large workloads.
  version: '1.0'
  contact:
    name: Jina AI
    url: https://jina.ai
servers:
- url: https://api.jina.ai/v1
  description: Jina AI production API
security:
- BearerAuth: []
tags:
- name: Reader
  description: URL-to-markdown extraction
paths:
  /:
    post:
      tags:
      - Reader
      summary: Read a URL as LLM-friendly content
      description: Submit a URL to extract clean, structured content suitable for LLM consumption. Output format is controlled by the X-Return-Format header.
      operationId: readUrl
      parameters:
      - name: X-Engine
        in: header
        schema:
          type: string
          enum:
          - browser
          - direct
          - cf-browser-rendering
        description: Rendering engine
      - name: X-Return-Format
        in: header
        schema:
          type: string
          enum:
          - markdown
          - html
          - text
          - screenshot
          - pageshot
        description: Desired output format
      - name: X-With-Links-Summary
        in: header
        schema:
          type: boolean
      - name: X-With-Images-Summary
        in: header
        schema:
          type: boolean
      - name: X-With-Generated-Alt
        in: header
        schema:
          type: boolean
      - name: X-Timeout
        in: header
        schema:
          type: integer
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/ReaderRequest'
      responses:
        '200':
          description: Content extracted successfully
          content:
            text/plain:
              schema:
                type: string
            application/json:
              schema:
                $ref: '#/components/schemas/ReaderResponse'
        '401':
          description: Unauthorized
  /{url}:
    get:
      tags:
      - Reader
      summary: Read a URL via path
      description: Convenience GET form that accepts the target URL as a path segment.
      operationId: readUrlByPath
      parameters:
      - name: url
        in: path
        required: true
        schema:
          type: string
        description: Target URL to extract
      responses:
        '200':
          description: Extracted content
          content:
            text/plain:
              schema:
                type: string
components:
  schemas:
    ReaderResponse:
      type: object
      properties:
        code:
          type: integer
        status:
          type: integer
        data:
          type: object
          properties:
            title:
              type: string
            description:
              type: string
            url:
              type: string
            content:
              type: string
            usage:
              type: object
              properties:
                tokens:
                  type: integer
    ReaderRequest:
      type: object
      required:
      - url
      properties:
        url:
          type: string
          format: uri
          example: https://example.com
        viewport:
          type: object
          properties:
            width:
              type: integer
            height:
              type: integer
        injectPageScript:
          type: string
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: API Key