Custom Document Extraction
A 3-step recipe with Mindee. We can help you get access.
Recipe · 3 steps
Custom Document Extraction
with Mindee
Upload a custom document with raw text capture and read fields plus full text.
- Send the document to the asynchronous extraction queue against the custom model with raw_text enabled so the full document text is returned.
- Poll the shared jobs endpoint until the custom extraction job reports Processed or Failed.
- Retrieve the completed extraction inference and read the custom fields and the full raw text parsed from the document.
What you'll need to have ready
- authorization: Mindee API key sent in the Authorization header.
- modelId: UUID of the custom extraction model to apply.
- file: The document file to upload as binary form data.
- filename: Optional filename to associate with the uploaded document.
- textContext: Optional additional context passed to the model for this inference.
For your developer
Arazzo workflow mindee:custom-document-extraction, 3 operations: Enqueue_Extraction_Product_Inference_v2_products_extraction_enqueue_post → Get_Job_Status_v2_jobs__job_id__get → Get_Extraction_Product_Result_v2_products_extraction_results__inference_id__get. Specs on APIs.io.