Unstract
No-code platform turning PDFs, scans and images into structured JSON with LLMs. Define the schema in plain English, deploy as a REST API or ETL pipeline.
// OCR
9 entries tagged with "ocr"
// ENTRIES
9 entries
Chandra
OCR model that turns images and PDFs into Markdown, HTML or JSON while keeping layout. Handles tables, math, handwriting and forms across 90+ languages.
Capso
Native macOS screenshot and screen recording app in the vein of CleanShot X. Annotation, OCR, webcam recording, a video editor and capture history built in.
PaddleOCR
Open-source OCR toolkit that turns images and PDFs into structured, LLM-ready data (Markdown, JSON). Handles 100+ languages, tables, formulas and charts.
Receipt OCR
Python OCR engine for receipts: raw text extraction via Tesseract and structured data via LLMs (OpenAI, Gemini, Groq). Ships with a CLI and FastAPI.
Falcon-Perception
PyTorch inference engine for vision-language models: detection, segmentation and OCR driven by natural language queries, with paged KV cache and CUDA graphs.
MarkItDown
Microsoft's Python library and CLI to convert PDF, Word, Excel, images, audio and more into clean Markdown, ready for LLMs and text analysis pipelines.
Stirling PDF
Self-hosted open-source PDF toolkit with 50+ built-in tools. Merge, split, OCR, sign, convert and compress PDFs with REST API and workflow automation.
Tesseract.js
JavaScript OCR library that runs in the browser and Node.js. Recognizes text in 100+ languages via WebAssembly with no server-side processing needed.