Papra
Keep invoices and contracts on your own server and find them by their text: OCR, email or watched-folder intake, one Docker container running on SQLite.
// OCR
11 entries tagged with "ocr"
// COMPARISONS
1// ENTRIES
10 entries
Unstract
Turn invoices, contracts and scans into JSON with an LLM, as an API or ETL pipeline. A self-hosted AGPL platform that hands OCR to a separate text extractor.
Chandra
Scans and PDFs turned into structured Markdown by a vision model, handwriting included. Free commercial use of the weights ends above $2M in revenue or funding.
Capso
Annotated screenshots, MP4 or GIF recordings and OCR on macOS 15 and later: a native Swift app pitched as a free CleanShot X alternative, under BSL 1.1.
PaddleOCR
Pull text, tables and formulas out of PDFs and images as Markdown or JSON for an LLM: OCR and vision models under Apache 2.0, on PaddlePaddle by default.
Receipt OCR
Turns a photo of a receipt into JSON: merchant, date, line items, total. A Python library that hands the image to an OpenAI-compatible LLM. API key needed.
Falcon-Perception
Pick out objects from a text prompt or OCR a whole page: the inference code for TII's Falcon vision-language models, built for NVIDIA GPUs and Apple Silicon.
MarkItDown
Word, Excel, PowerPoint, PDF and EPUB files turned into Markdown an LLM can read, in Python. Images and scanned PDFs only become text by calling an LLM.
Stirling PDF
Sign, redact, OCR and convert PDFs on your own machine or server instead of an online service. Open core, free up to five users, with automation capped monthly.
Tesseract.js
Read text from images in the browser or Node.js without an OCR server: the Tesseract engine compiled to WebAssembly. Printed text only, and no PDF input.