Skip to content

Changelog

PyPI

Warning

RapidOCR API now has its own repository, so this changelog is no longer updated. See the latest releases at RapidOCR API releases.

2025-01-01 v0.1.5 update

Merged PR #309.

  • Added timestamps to Uvicorn logs.
  • Added Vim to the Docker image for temporary editing.
  • Added a timezone environment variable when running Docker so timestamps use local time.

2024-12-04 v0.1.4 update

Merged PRs #282 and #281. Added support for selecting among multiple RapidOCR inference engines.

2024-11-12 v0.1.2 update

  • Merged PR #263: fixed an incorrect floating-point type that prevented scores from being displayed.

2024-10-28 v0.1.1 update

  • Removed the mistakenly used async declaration from def ocr(). It may be added again in the future if needed.

2024-10-28 v0.1.0 update

🍿2024-10-15 v0.0.9 update

2024-07-11 v0.0.7 update

🥥2024-03-04 v0.0.6 update

  • Improved image loading to match rapidocr_onnxruntime>=1.3.13.
  • Unified API return values as dictionaries.

🍜2023-05-22 API update

  • Decoupled the API from OCRWeb as a separate module; see the API source.
  • After rapidocr_web>0.1.6, install the API directly with pip install rapidocr_api instead of pip install rapidocr_web[api].

Comments