Paperwise

Paperwise turns PDFs, scans and records into a searchable library for people managing private documents. It answers questions across files and links its answers to the source pages so you can check the evidence.

• Document reading: Use local Tesseract or LLM OCR to extract text from scans and image-heavy documents, with support for screenshots and text dumps alongside PDFs.
• Automatic organization: Classify documents by type, correspondent and date using a taxonomy you control, then refine rules or merge categories as the library grows.
• Cross-document answers: Compare records, reconcile bills with insurance statements or collect monthly usage into a table with page-level citations.
• Separate model choices: Assign different models to OCR, metadata extraction and Q&A, using self-hosted models, OpenAI, Gemini or an OpenAI-compatible API.
• Automatic intake and mobile access: A watched folder feeds files into OCR, extraction, tagging and indexing; Paperless Mobile can browse, search, filter and upload through a compatible API.

Paperwise runs on your machine or server through Docker Compose and stores uploaded files and extracted data locally. Local Tesseract keeps OCR on your hardware. Hosted models receive the document content needed for their tasks, so keeping AI processing local depends on your backend choices. The app has no telemetry.

Paperless Mobile compatibility covers core document workflows rather than the full Paperless-ngx API. Document deletion is permanent, with no trash retention or restore.