mindee/doctr

doctr

✅ Recently maintained

docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.

6,406Stars
690Forks
+107-day stars

Python · Apache-2.0 · Last push 2026-10-31

Official repository & installation ↗
ocrdeep-learningdocument-recognitiontext-detectiontext-recognitionpytorchdocumentailayout-analysistable-structure-recognitiondocument-parsing

Star history

2025-09-292026-11-02
View historical data
DateStars
2026-10-046,367
2026-10-056,370
2026-10-066,370
2026-10-076,372
2026-10-086,374
2026-10-096,376
2026-10-106,378
2026-10-116,378
2026-10-126,379
2026-10-136,381
2026-10-146,382
2026-10-156,386
2026-10-166,387
2026-10-176,389
2026-10-186,390
2026-10-196,390
2026-10-206,390
2026-10-216,391
2026-10-226,392
2026-10-236,393
2026-10-246,394
2026-10-256,396
2026-10-266,397
2026-10-276,399
2026-10-286,402
2026-10-296,404
2026-10-306,405
2026-10-316,407
2026-11-016,405
2026-11-026,406

Get the project

git clone https://github.com/mindee/doctr.git

Related projects & alternative candidates

Related domains do not imply feature parity. Review official documentation and licenses.

LLMs-from-scratch

Step-by-step guide to building a GPT-like LLM from scratch in PyTorch.

★ 107,671PaddleOCR

Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports

★ 92,095MinerU

High-precision PDF and academic document parser to structured Markdown.

★ 82,969