Open-source alternatives to Amazon Textract
4 open-source projects that do what Amazon Textract does. Picked or featured by Willet; numbers are from GitHub.
Last updated
| Project | What it does | Stars | License | Last commit | Self-hostable |
|---|---|---|---|---|---|
| OCR toolkit that converts scanned pages and photos to Markdown or JSON | 91.4k | Apache-2.0 | No | ||
| Python library and CLI that converts documents into structured data for AI | 69.2k | MIT | No | ||
| OCR model converting scanned pages and PDFs into layout-aware Markdown | 12.5k | Apache-2.0 | No | ||
| Self-hosted platform for extracting structured JSON from documents with LLMs | 7.3k | AGPL-3.0 | Yes |