Parse Document

POST

/documents/parse

curl --request POST \
  --url https://example.com/documents/parse \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form file=example \
  --form backend=hybrid-auto-engine \
  --form parse_method=auto \
  --form language=en \
  --form formula_enable=true \
  --form table_enable=true

Submit a document for parsing via MinerU and return immediately with a task_id.

MinerU provides full document understanding: layout analysis, OCR, table/formula extraction, and structured markdown output. Accepts PDF, images, and DOCX files.

Poll GET /documents/parse/{task_id} for the result.

Authorizations

• HTTPBearer

Request Body^required

Media typemultipart/form-data

Body_parse_document_documents_parse_post

object

file

required

File

PDF, image, or DOCX file to parse

string

backend

Backend

MinerU backend: hybrid-auto-engine, pipeline, vlm-auto-engine

string

default: hybrid-auto-engine

parse_method

Parse Method

Parse method: auto, txt, ocr

string

default: auto

language

Language

OCR language hint

string

default: en

formula_enable

Formula Enable

Enable formula parsing

boolean

default: true

table_enable

Table Enable

Enable table parsing

boolean

default: true

Responses

202

Successful Response

Media typeapplication/json

DocumentParseSubmission

object

task_id

required

Task Id

Celery task ID for tracking the document parse job

string

status

Status

Initial status of the task

string

default: queued

Example

{
  "status": "queued"
}

422

Validation Error

Media typeapplication/json

HTTPValidationError

object

detail

Detail

Array<object>

ValidationError

object

loc

required

Location

Array

msg

required

Message

string

type

required

Error Type

string

input

Input

ctx

Context

object

Example^generated

{
  "detail": [
    {
      "loc": [
        "example"
      ],
      "msg": "example",
      "type": "example",
      "input": "example",
      "ctx": {}
    }
  ]
}