# RapidAI/RapidOCR

📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.

Repository: https://github.com/RapidAI/RapidOCR
Canonical: https://ross.abutalabs.com/products/rapidocr
Homepage: https://rapidai.github.io/RapidOCRDocs
Language: Python
License: Apache-2.0
License Family: permissive
Topics: ocr, onnxruntime, crnn, dbnet, openvino, rapidocr, chineseocr, easyocr, paddleocr, onnxocr, mnn, tensorrt
Last push: 2026-08-26T01:40:46+00:00

## Health v2 (maintenance only)
Score: 97/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 94, longevity 100
- inputs: {"age_days": 2067, "days_push": 8, "days_rel": 43, "gap_med": 13.5, "n_releases_24m": 35}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7599, forks 708 (observed 2026-08-28T04:10:01.927568+00:00)

## What it is
RapidOCR is an open-source, multi-language OCR toolkit that performs text detection and recognition using models converted to run on ONNX Runtime, OpenVINO, MNN, TensorRT, and other inference engines. It supports fast offline deployment across Linux, Windows, and macOS with default Chinese and English recognition and additional language models available.

## Use cases
- extract text from images in python
- chinese and english ocr offline
- convert scanned pdfs to text
- run ocr without gpu or cloud services
- deploy text recognition on multiple platforms
- fast ocr for screenshots and documents

## When to choose
- you need free, offline OCR with no cloud dependency
- you need Chinese and English text recognition out of the box
- you want fast inference via ONNX Runtime or OpenVINO across platforms

## When to avoid
- you need handwriting or rare-language recognition not covered by the model list
- you require training custom OCR models from scratch
- you need a managed cloud OCR API with enterprise support

## Facets
- artifact type: library
- maturity: active
- function: ocr, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, developer-tools
- platform: python, windows, cross-platform
- tags: text-detection, text-recognition, onnxruntime, openvino, paddleocr, chinese-ocr, offline-deployment, natural-language-processing, linux, macos

## Member repositories
- RapidAI/RapidOCR (main) score 97

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:01.927568+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:36:51.484180+00:00, confidence not recorded.
  - readme: https://github.com/RapidAI/RapidOCR (fetched 2026-08-28T04:10:01.927568+00:00, sha 1d2e1bc0cf2a)
  - homepage: https://rapidai.github.io/RapidOCRDocs (fetched 2026-08-29T08:32:06.555464+00:00, sha 7d7caab70f34)
  - registry_pypi: https://pypi.org/pypi/rapidocr/json (fetched 2026-08-29T08:32:06.557840+00:00, sha 9f03d0ae75dc)
- Data as of 2026-08-30T08:39:29.467469+00:00.