---
title: "Start a parse job"
method: POST
path: "/v1/parsing/jobs"
tags: ["parsing"]
---

# Start a parse job

`POST /v1/parsing/jobs`

Start a parse job for the provided file.

Args:
    params: The parameters for creating a parse job.

Returns:
    The created parsing job.

## Request body

- ParsingJobCreateParams — Parameters for creating a parsing job.
  - `file_id` string, required — The ID of the file to parse
  - `element_types` ElementType[], nullable — The elements to extract from the document
  - `chunking_strategy` 'page' — Strategy used for chunking document content.
  - `return_format` 'html' | 'markdown' | 'plain' — Format options for the returned document content.
  - `mode` 'fast' | 'high_quality' — The mode of the document parser.

## Response `201`

The created parse job

- ParsingJobResponse — A parsing job with its result narrowed to the public fields.
  - `id` string, required — The ID of the job
  - `file_id` string, required — The ID of the file to parse
  - `filename` string, nullable — The name of the file
  - `status` 'pending' | 'in_progress' | 'cancelled' | 'completed' | 'failed', required
  - `error` object, nullable — The error of the job
  - `result` DocumentParserResultResponse — The public result of a parsing job.
    - `chunking_strategy` 'page', required — Strategy used for chunking document content.
    - `return_format` 'html' | 'markdown' | 'plain', required — Format options for the returned document content.
    - `element_types` ElementType[], required — The types of elements extracted
    - `chunks` ChunkResponse[], required — List of extracted chunks from the document
      - `content` string, nullable — The full content of the chunk
      - `content_to_embed` string, required — The content of the chunk to embed
      - `elements` ChunkElementResponse[], required — List of elements contained in this chunk
        - `type` 'header' | 'footer' | 'title' | 'section-header' | 'page-number' | 'list-item' | 'figure' | 'table' | 'form' | 'text' | 'footnote', required — Types of elements that can be extracted from a document.
        - `confidence` number, required — The confidence score of the extraction
        - `bbox` unknown[], required — The bounding box coordinates [x1, y1, x2, y2]
          - unknown
        - `page` integer, required — The page number where the element was found
        - `content` string, required — The extracted text content of the element
        - `summary` string, nullable — A brief summary of the element's content
        - `image` string, nullable — The base64-encoded image data for figure elements
    - `page_sizes` array[] — List of (width, height) tuples for each page
      - unknown[]
        - unknown
  - `started_at` string, date-time, nullable — The started time of the job
  - `finished_at` string, date-time, nullable — The finished time of the job
  - `created_at` string, date-time — The creation time of the job
  - `updated_at` string, date-time, nullable — The updated time of the job
  - `object` 'parsing_job' — The type of the object

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/mixedbread/apis/mxbai-omni.md) · [All operations](https://skmtc.net/mixedbread/apis/mxbai-omni/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/mixedbread/mxbai-omni/revisions/6dac6bb49f72/schema)