> ## Documentation Index
> Fetch the complete documentation index at: https://help.abacusdocs.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Create a batch job

> Start a job from Batch Processing. Pick a schema and add up to 1,024 files.

The sidebar label is **Batch Processing**. The page title is **Intelligent Document Processing** (`/idp/`).

Use a batch when the same [schema](/schemas/overview) should run across many files. Single-file work stays on [My Documents](/guides/uploading-documents).

## Create a job

1. Open **Batch Processing**.
2. Start a **New Batch Job**.
3. Name the batch and pick a **default schema** (a published version).
4. Add files or folders. Images in a folder become one PDF.
5. Submit. The batch moves **Created → Queued → Processing → Completed** (or **Failed**).

Limits:

* Up to **1,024** files per batch.
* Each file up to **64 MB**.
* Same formats as My Documents: PDF and images only. [Digital uploads](/digital-documents) are not available yet.

## Progress

Open the batch to watch per-document status. A partial batch is normal: completed documents are not re-run when you retry failures.

You can **cancel** a running batch. Canceled batches cannot receive new files.

## After it finishes

* Open the [results grid](/guides/exporting-results) to edit cells and export.
* **Open in chat** creates an Extract Chat thread with this batch attached.
* **Share** to the [Org Library](/sharing/org-library) or via Direct Share.
* Reprocess selected documents, or [move a document](/guides/exporting-results#move-a-document) into another batch.

<Note>
  Credits are charged per document as it completes, attributed to the person who started or retried that file. The organization pays; the usage table names the actor.
</Note>