> ## Documentation Index > Fetch the complete documentation index at: https://docs.landing.ai/llms.txt > Use this file to discover all available pages before exploring further. # ADE Split > Split a Markdown document into segments and return the split response inline. ## OpenAPI ````yaml /dpt3/openapi-adev2.json post /v1/split openapi: 3.1.0 info: title: >- LandingAI Agentic Document Extraction (ADE) Gen2 API: Parse, Extract, and Ground Documents with DPT-3 description: >- Parse and extract data from documents with DPT-3, a document parsing model from LandingAI, using the Agentic Document Extraction (ADE) v2 endpoints. Parse converts PDFs and images into structured Markdown and elements with per-element grounding (page numbers and coordinates) for RAG, search, and extraction pipelines. Extract pulls specific fields from parsed Markdown using a JSON schema. Ground maps extracted fields back to the source blocks, with page numbers and bounding boxes for each value. Documentation: https://docs.landing.ai version: 1.0.0 servers: - url: https://api.ade.landing.ai description: Production API security: [] paths: /v1/split: post: tags: - Split summary: ADE Split description: >- Split a Markdown document into segments and return the split response inline. operationId: v1-split_run_sync requestBody: required: true content: multipart/form-data: schema: type: object required: - split_class properties: markdown: anyOf: - type: string format: binary - type: string description: >- Markdown content to split, as an inline string or an uploaded file. Provide either `markdown` or `markdown_url`, not both. markdown_url: type: string description: >- A publicly accessible URL to the Markdown file to split. Provide either `markdown` or `markdown_url`, not both. split_class: type: string description: >- The split classification entries, as a JSON-encoded array of objects with `name` (required), `description`, and `identifier` keys. At most 19 entries. Sent as a JSON-serialized string in form data. model: type: string description: >- The split model version to use. Accepts `split-latest`, `split`, or a dated snapshot. Defaults to the latest snapshot. The resolved version is echoed back as `metadata.version`. responses: '200': description: The split response content: application/json: schema: $ref: '#/components/schemas/SplitResponse' '422': description: Request validation failed. content: application/json: schema: $ref: '#/components/schemas/V1LegacyErrorResponse' x-codeSamples: - lang: cURL label: cURL source: |- curl -X POST 'https://api.ade.landing.ai/v1/split' \ -H 'Authorization: Bearer YOUR_API_KEY' \ -F 'markdown=@parse-output.md' \ -F 'split_class=[{"name":"Bank Statement","description":"Document from a bank that summarizes all account activity over a period of time."},{"name":"Pay Stub","description":"Document that details an employee'\''s earnings, deductions, and net pay for a specific pay period.","identifier":"Pay Stub Date"}]' - lang: Python label: Python source: >- import json import requests url = 'https://api.ade.landing.ai/v1/split' headers = {'Authorization': 'Bearer YOUR_API_KEY'} files = {'markdown': open('parse-output.md', 'rb')} data = {'split_class': json.dumps([{"name": "Bank Statement", "description": "Document from a bank that summarizes all account activity over a period of time."}, {"name": "Pay Stub", "description": "Document that details an employee's earnings, deductions, and net pay for a specific pay period.", "identifier": "Pay Stub Date"}])} response = requests.post(url, headers=headers, files=files, data=data) print(response.json()) - lang: JavaScript label: Node.js source: >- const axios = require('axios'); const FormData = require('form-data'); const fs = require('fs'); const form = new FormData(); form.append('markdown', fs.createReadStream('parse-output.md')); form.append('split_class', JSON.stringify([{"name": "Bank Statement", "description": "Document from a bank that summarizes all account activity over a period of time."}, {"name": "Pay Stub", "description": "Document that details an employee's earnings, deductions, and net pay for a specific pay period.", "identifier": "Pay Stub Date"}])); axios.post('https://api.ade.landing.ai/v1/split', form, { headers: { 'Authorization': 'Bearer YOUR_API_KEY', ...form.getHeaders() } }) .then(response => console.log(response.data)) .catch(error => console.error(error)); components: schemas: SplitResponse: description: 'The split result: the merged `splits` segments and request `metadata`.' properties: splits: description: >- The split segments, in page order. Consecutive pages with the same classification merge into one segment; an identifier change starts a new segment. items: $ref: '#/components/schemas/Split' title: Splits type: array metadata: $ref: '#/components/schemas/SplitMetadata' description: >- Information about the request: file name, page count, duration, credits, and the resolved model version. required: - splits - metadata title: SplitResponse type: object V1LegacyErrorResponse: title: V1LegacyErrorResponse description: >- The v1 error envelope: `message` for a request-validation failure and for an authentication, payment or permission rejection (401/402/403), `error` for every other rejection, never both. `code` is always present. oneOf: - type: object required: - code - message properties: code: type: string description: Stable snake_case error code. message: type: string description: Human-readable detail. - type: object required: - code - error properties: code: type: string description: Stable snake_case error code. error: type: string description: Human-readable detail. Split: description: |- One split segment: consecutive pages of one classification (an identifier change starts a new segment). properties: classification: description: The split classification name this segment was assigned. title: Classification type: string identifier: anyOf: - type: string - type: 'null' description: >- The identifier value extracted for this segment, when the matching split classification requested one. Null otherwise. title: Identifier pages: description: 0-indexed page numbers belonging to this segment, in order. items: type: integer title: Pages type: array markdowns: description: The Markdown content of each page in this segment, in order. items: type: string title: Markdowns type: array required: - classification - identifier - pages - markdowns title: Split type: object SplitMetadata: description: Information about a split request. properties: filename: description: >- Display name of the split document: the uploaded file's name for `markdown` file uploads (`.md` is appended when the name has no suffix), the URL path's file name for `markdown_url` inputs, or a generated name for inline Markdown. title: Filename type: string org_id: anyOf: - type: string - type: 'null' description: Organization ID. title: Org Id page_count: description: Total number of pages in the input Markdown. title: Page Count type: integer duration_ms: description: Total processing time in milliseconds. title: Duration Ms type: integer credit_usage: description: >- Credits consumed by this request: the input Markdown length in characters divided by 5000. Billed usage rounds this up to the next 0.1 credit. title: Credit Usage type: number job_id: description: >- The split job identifier for this request. When present it is the id of the request's entry in your billing dashboard — whichever stack served it — so it is the one to quote for this run. Empty when the serving stack reported no id, in which case there is no billing entry to quote. title: Job Id type: string version: description: The exact split model snapshot that processed the document. title: Version type: string required: - filename - org_id - page_count - duration_ms - credit_usage - job_id - version title: SplitMetadata type: object ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.