Unstructured
Bundles contain custom components that support specific third-party integrations with Orbis Studio.
This page describes the components that are available in the Unstructured bundle.
Unstructured API
The Unstructured API component parses documents (PDF, DOCX, HTML, images, and more) using Unstructured.io's serverless API and returns the extracted content as structured Data.
It supports multiple chunking strategies for downstream RAG workflows.
Unstructured API parameters
Some parameters are hidden by default in the visual editor. You can modify all component parameters through the component inspection panel that appears when you select a component.
| Name | Type | Description |
|---|---|---|
| api_key | SecretString | Input parameter. Your Unstructured.io serverless API key. Create one at app.unstructured.io. |
| api_url | String | Input parameter. The Unstructured API endpoint URL. Default: https://api.unstructuredapp.io/general/v0/general. |
| file | File | Input parameter. The file to parse. Supports PDF, DOCX, PPTX, HTML, images, and more. |
| chunking_strategy | String | Input parameter. How to chunk the extracted content. Options: None, basic, by_title, by_page, by_similarity. See Unstructured chunking docs. |
| unstructured_args | Dict | Input parameter. Additional arguments to pass to the Unstructured API. |
Unstructured API outputs
| Name | Type | Description |
|---|---|---|
| data | JSON | Output parameter. Parsed and chunked document content as Data records. |