View only. Sign in to run this demo. You can also run it with your own model key, without an account.
Extractor
Automatically find and list important details from any document
How to use the Extractor
- 1Add the documentUpload a PDF, Word file or image, paste text, or start from a ready-made document.
- 2Choose how it is readFor an uploaded file, pick the parser that reads it. Auto-detect chooses by file type.
- 3Say what to extractDescribe it in words and generate the schema, or list the fields yourself.
- 4Review the resultEvery field is shown, also those with no value. Copy or download the JSON.
- Trace sources asks the model for the exact text each value came from. The quote is checked against the text and, for a PDF, located and highlighted on its page, so a value can be checked at a glance.
- Auto-detect reads a PDF with PyMuPDF, a Word file with DOCX and an image with Vision. A scanned PDF has no text layer: choose Vision, Vision+, Multimodal or the HH Parser.
- Nothing you enter is saved. The examples come with a ready-made schema. A schema generated for your own task, or for an edited example, lives in this session only and is used again while the task stays the same.
Start from a ready-made document
1. The document
Upload a file and choose how it is read, or paste the text.
2. What to extract
Describe the data in words, or list the fields yourself.
This task has a ready-made schema, so only the extraction runs. Change the text and a new schema is generated.
Each value comes with its quote, highlighted in the PDF. Uses more output tokens.
Model: the server's model
No extraction yet
Add a document and press Extract data. Every field of the result is shown here, also those with no value.
