Response Format

A split run returns one entry per detected sub-document, each with its type, page range, an optional extracted identifier, and a standalone fileId you can pass into other endpoints. This page explains every field in the response.

Response structure

A completed split run looks like this (truncated to two splits for brevity). The result lives in output.splits.

1 {
2   "object": "split_run",
3   "id": "splr_Xj8mK2pL9nR4vT7qY5wZ",
4   "status": "PROCESSED",
5   "file": {
6     "object": "file",
7     "id": "file_GzKUy0VDhHscv7tweODYb",
8     "name": "loan_application.pdf"
9   },
10   "output": {
11     "splits": [
12       {
13         "id": "splt_xK9mLPqRtN3vS8wF5hB2cQ",
14         "classificationId": "loan_application",
15         "type": "loan_application",
16         "startPage": 1,
17         "endPage": 5,
18         "identifier": "Jordan Avery",
19         "observation": "Pages 1-5 are a Uniform Residential Loan Application.",
20         "fileId": "file_8sLPqRtN3vS2wF5hB2cQ"
21       },
22       {
23         "id": "splt_2pL9nR4vT7qY5wZj8mK2",
24         "classificationId": "bank_statement",
25         "type": "bank_statement",
26         "startPage": 6,
27         "endPage": 9,
28         "identifier": "Jordan Avery",
29         "observation": "Pages 6-9 are a monthly bank statement.",
30         "fileId": "file_R4vT7qY5wZj8mK2pL9nR"
31       }
32     ]
33   },
34   "reviewed": false,
35   "edited": false,
36   "usage": { "credits": 3 }
37 }

Top-level fields

Field	Type	Description
`object`	string	Always `"split_run"`.
`id`	string	Unique identifier for the run (e.g. `splr_...`). Use it to fetch results later.
`file`	object	The processed file (`id`, `name`). Reusable as input to other endpoints.
`status`	string	`PENDING`, `PROCESSING`, `PROCESSED`, `FAILED`, or `CANCELLED`.
`output`	object \| null	The split result. Present when `status` is `PROCESSED`. Contains `splits`.
`reviewed` / `edited`	boolean	Whether a human reviewed the run, and whether they changed the result. When `reviewed` is `true`, `initialOutput` and `reviewedOutput` are also populated.
`splitter` / `splitterVersion`	object \| null	The splitter and version used for the run. Present when a splitter reference was provided; not present when using inline `config`.
`config`	object	The full configuration used, including defaults that were applied.
`parseRunId`	string \| null	The ID of the parse run used for this split run.
`dashboardUrl`	string	Link to view the run in the Extend dashboard.
`usage`	object	Credits consumed (`usage.credits`).
`failureReason` / `failureMessage`	string \| null	Machine-readable code and human-readable message. Present when `status` is `FAILED`.

The splits array

output.splits contains one object per detected sub-document, in document order. Each split describes a contiguous page range and the type the splitter assigned to it.

1 {
2   "id": "splt_xK9mLPqRtN3vS8wF5hB2cQ",
3   "classificationId": "loan_application",
4   "type": "loan_application",
5   "startPage": 1,
6   "endPage": 5,
7   "identifier": "Jordan Avery",
8   "observation": "Pages 1-5 are a Uniform Residential Loan Application.",
9   "fileId": "file_8sLPqRtN3vS2wF5hB2cQ"
10 }

Split fields

Field	Type	Description
`id`	string	Unique ID for this split.
`classificationId`	string	The `id` of the classification that matched (set in the config). Branch your logic on this — it’s stable, unlike `type`.
`type`	string	The document type, corresponding to the `classificationId`.
`startPage`	integer	The start page of the sub-document.
`endPage`	integer	The end page of the sub-document.
`identifier`	string	The extracted identifier (e.g. an invoice number), driven by the classification’s `identifierKey`.
`observation`	string	A short explanation of why these pages were grouped as this type.
`fileId`	string	The file ID of the standalone sub-document, usable as input to other endpoints.
`name`	string	Optional name for the split.

The fileId on each split points to a real, separate file. The common pattern is to split once, then feed each split’s fileId into Parse, Extract, or a Workflow.