Parse & Extract Data from PDFs/Images with Mistral OCR

This n8n workflow leverages Mistral OCR API to parse and extract structured data from documents like multi-page PDFs or single images. It handles bank statements and screenshots by sending HTTP requests to the API, supporting direct uploads or secure methods via Mistral Cloud storage with signed URLs to prevent unauthorized access. Unique features include automatic document page splitting and markdown conversion, reducing reliance on tools like StirlingPDF.

Key benefits include high accuracy in table extraction, AI-powered document understanding for querying content (e.g., counting deposits in statements), and flexibility for public or private files from AWS S3 or Cloudflare R2. It's ideal for automating data processing in finance, health records, or any document-heavy workflows, saving hours of manual effort.

Use cases span financial analysis (bank statements), compliance reporting, invoice processing, and Q&A on docs. Requires a Mistral Cloud API key; workflow demonstrates three methods: direct OCR, secure upload, and intelligent querying with Mistral-small.