Skip to main content
This quickstart uses Unstructured Transform, which works only with local files.To work with remote files and data, skip this quickstart and use the Unstructured Pipelines API instead.

Requirements

1

Sign up

  1. Go to https://transform.unstructured.io.
  2. Follow the on-screen instructions to sign up.
2

Get your API key and API URL

  1. Sign in to https://transform.unstructured.io, if you are not already signed in.
  2. On the sidebar, click Get started.
  3. In the Get your API key and URL area, get the value of your Unstructured API key. You can click the pages icon (Copy) to copy this value to your system’s clipboard, or click the eyes icon to show this value so you can manually select and then copy it.
  4. In this same area, get the value of your Unstructured API URL, which should be https://platform-api.transform.unstructured.io/api/v1
3

Set environment variables

This quickstart assumes that you have set two local environment variables that your local curl scripts or Python code can access. These two environment variables are:
  • UNSTRUCTURED_API_URL, representing your Unstructured API URL, which you got from the previous step.
  • UNSTRUCTURED_API_KEY, representing your Unstructured API key, which you got from the previous step.
For instructions about how to set these environment variables accordingly, see your local operating system’s documentation.
4

Install the tools

For curl users: To use this quickstart with curl, you only need to have curl installed locally. For installation instructions, see your local operating system’s documentation.For Python users: To use this quickstart with Python, install the Unstructured Python SDK locally by running the following command:
pip install unstructured-client

Quickstarts by use case

Transform

Transform your documents’ contents into text and metadata, which Unstructured outputs as standard AI-ready JSON.

Enrich

Uncover more context and insights from your documents, which Unstructured adds to your existing AI-ready JSON.

Embed

Power up semantic search and retrieval with vector embeddings that Unstructured generates. These embeddings are based on your documents’ text content.

Extract

Extract just the data that you care about from your documents, which Unstructured outputs as JSON object data records.