Indra OCR

Use it locally to extract and structure text from scanned documents and images.

Availability

Install with pip install indra-ocr; model weights available on Hugging Face

Pricing model

Free for research, personal use, and startups under $20M funding/revenue; broader commercial licensing available

Deployment

Runs on CPU; no GPU required

Privacy

Local viewer files stay on your machine

Licence

Code: Apache 2.0; model weights: modified OpenRAIL-M

• Indra OCR converts PDFs and images into Markdown, text, or JSON with boxes. ^{[1] [2]}

  It turns PDFs and images into markdown, text, or JSON with boxes, and it reads handwriting, 80+ languages, and pages at any angle.

• It supports more than 80 languages and handwritten pages. ^{[1] [2]}

  - **80+ languages** - Latin-script languages, Russian and the Cyrillic languages, Greek, Chinese, Japanese, Korean, Arabic, Persian, Urdu, Hebrew, Hindi and the Indic languages, Thai, Khmer and more ([languages](#languages))

• It runs on CPU without PyTorch or a GPU. ^{[1] [2]}
• The package can be installed with pip, and the models are available on Hugging Face. ^{[1] [2]}

  The models download from Hugging Face on the first run and are cached.

• The model weights are free for research, personal use, and startups under $20M funding or revenue. ^{[1] [2]}

  This code is Apache 2.0, and our model weights use a modified OpenRAIL-M license (free for research, personal use, and startups under $20M funding/revenue, cannot be used competitively with our API).

• Extract structured text from PDFs and images on a CPU Daily brief 2026-10-14