> For the complete documentation index, see [/llms.txt](/llms.txt).

> Source: /api/information-extraction/universal-extraction/async

# Asynchronous API

## Create Async Job

Submit multiple documents for asynchronous information extraction processing.

OpenAPI spec: `./openapi/openapi-information-extract.yaml`

- `POST /information-extraction/async`

<br />

### Response

**Return values**

Returns a job creation response object.

**The job creation object**

<table><tbody><tr><td><strong>job\_id</strong> string<br />UUID-based unique identifier for the created job.</td></tr><tr><td><strong>status</strong> string<br />Current job status. Can be `CREATED`, `IN_PROGRESS`, `PARTIALLY_COMPLETED`, `COMPLETED`, `FAILED`, or `CANCELED`.</td></tr><tr><td><strong>created\_at</strong> string<br />Job creation timestamp in UTC (ISO 8601 format).</td></tr><tr><td><strong>message</strong> string<br />Success message confirming job creation and processing initiation.</td></tr></tbody></table>

---

## Retrieve Job

Get detailed information about a specific information extraction job, including processing status and results.

OpenAPI spec: `./openapi/openapi-information-extract.yaml`

- `GET /information-extraction/jobs/{job_id}`

<br />

### Response

**Return values**

Returns a detailed job information object.

**The job object**

<table><tbody><tr><td><strong>job\_id</strong> string<br />UUID-based unique identifier for the job.</td></tr><tr><td><strong>status</strong> string<br />Current job status. Can be `CREATED`, `IN_PROGRESS`, `PARTIALLY_COMPLETED`, `COMPLETED`, `FAILED`, or `CANCELED`.</td></tr><tr><td><strong>model</strong> string<br />Information extraction model name used for processing.</td></tr><tr><td><strong>total\_documents</strong> integer<br />Total number of documents in the job.</td></tr><tr><td><strong>completed\_documents</strong> integer<br />Number of successfully processed documents.</td></tr><tr><td><strong>created\_at</strong> string<br />Job creation timestamp in UTC (ISO 8601 format).</td></tr><tr><td><strong>started\_at</strong> string<br />Job processing start timestamp in UTC (ISO 8601 format).</td></tr><tr><td><strong>completed\_at</strong> string<br />Job completion timestamp in UTC (ISO 8601 format). Present only when job is completed.</td></tr><tr><td><strong>failed\_at</strong> string<br />Job failure timestamp in UTC (ISO 8601 format). Present only when job has failed.</td></tr><tr><td><strong>cancelled\_at</strong> string<br />Job cancellation timestamp in UTC (ISO 8601 format). Present only when job is cancelled.</td></tr><tr><td><strong>error\_message</strong> string<br />Error message if job failed. Null if successful.</td></tr><tr><td><strong>schema</strong> object<br />Schema information used for extraction, including JSON schema definition.</td></tr><tr><td><strong>batches</strong> array<br />Array of per-document processing information. Includes batch ID, file name, status, and extraction results when `with_result=true`.</td></tr></tbody></table>

---

## List Jobs

Retrieve a paginated list of your information extraction jobs with filtering options.

OpenAPI spec: `./openapi/openapi-information-extract.yaml`

- `GET /information-extraction/jobs`

<br />

### Response

**Return values**

Returns a paginated list of job information objects.

**The jobs list object**

<table><tbody><tr><td><strong>total\_count</strong> integer<br />Total number of jobs available for the user.</td></tr><tr><td><strong>limit</strong> integer<br />Number of jobs returned per page (page size limit).</td></tr><tr><td><strong>offset</strong> integer<br />Starting position of current page (pagination offset).</td></tr><tr><td><strong>items</strong> array<br />Array of job information objects. Each item contains basic job details including job\_id, status, model, document counts, and timestamps.</td></tr><tr><td><strong>items\[].job\_id</strong> string<br />UUID-based unique identifier for each job.</td></tr><tr><td><strong>items\[].status</strong> string<br />Current job status. Can be `CREATED`, `IN_PROGRESS`, `PARTIALLY_COMPLETED`, `COMPLETED`, `FAILED`, or `CANCELED`.</td></tr><tr><td><strong>items\[].model</strong> string<br />Information extraction model name used for processing.</td></tr><tr><td><strong>items\[].total\_documents</strong> integer<br />Total number of documents in each job.</td></tr><tr><td><strong>items\[].completed\_documents</strong> integer<br />Number of successfully processed documents in each job.</td></tr><tr><td><strong>items\[].created\_at</strong> string<br />Job creation timestamp in UTC (ISO 8601 format).</td></tr><tr><td><strong>items\[].started\_at</strong> string<br />Job processing start timestamp in UTC (ISO 8601 format).</td></tr><tr><td><strong>items\[].completed\_at</strong> string<br />Job completion timestamp in UTC (ISO 8601 format). Present only when job is completed.</td></tr></tbody></table>

---

## Examples

Here's a typical workflow for using the async information extraction API:

### 1. Create Async Job

**Request**

_Request/response examples: information-extract-async / request / create_

**Response**

_Example: information-extract-async / response-200 / create / default_

### 2. Retrieve Job

**Request**

_Request/response examples: information-extract-async / request / retrieve_

**Response**

_Request/response examples: information-extract-async / response-200 / retrieve_

### 3. List Jobs

**Request**

_Request/response examples: information-extract-async / request / list_

**Response**

_Example: information-extract-async / response-200 / list / default_