- Explore demo pipelines – Open preloaded documents and see how parsing, extraction, and editing work.
- Upload your own files – Parse PDFs or scans and view the detected text, tables, and figures.
- Test extractions – Create or generate schemas and watch the output update in real time.
1. Get started
Visit studio.reducto.ai.You can explore demo pipelines immediately. To upload your own documents, sign up or log in.
If you’re part of an organization, your administrator must enable Studio access.
2. Try a Parse → Extract pipeline
Goal: Parse a document and extract structured data.- Open the “Financial Statement” demo.
-
Select the Parse tab. You’ll see the document with detected sections, bounding boxes, and figures.
Expected result: parsed content displayed alongside the original PDF.
-
Switch to the Extract tab. The schema and JSON output appear side by side.
Expected result: structured output linked to bounding boxes in the document.
Bounding boxes and output blocks are clickable — select either to highlight the matching content.
3. Try an Edit pipeline
Goal: Fill out a form automatically.- Open the “SS-4” demo.
- Select the Edit tab to see the instructions.
- Click Preview to view the filled-out form.
Expected result: a complete SS-4 form with data populated.
4. Upload your own file
Goal: Create your own pipeline and parse your own document.1
Sign up or log in
Sign up or log in to upload your own documents to Studio.
2
Create a Parse pipeline
Add a new pipeline, then choose Parse for processor type. Create a name, which can be edited later.


3
Upload a document
Click to upload or drag and drop a file to get started. We accept many filetypes, including PDF, DOCX, images, and more. You can download a sample PDF from a demo pipeline if you don’t have one on hand.


4
Click Run to parse
Running the Parse pipeline without any configuration adjustments will use our default settings. For more complex documents, check out our best practices for parsing guide.


5
Open the Result tab to view parsed text and structure
The results tab displays the original document with bounding box overlays with our Parse output side-by-side. You can switch between formatted markdown and the JSON response view, as well as copy the invocation code under the Code button.

