TL;DR

PaddleOCR now includes PaddleOCR-VL-1.6, a significantly improved document parsing model, offering state-of-the-art accuracy on benchmarks like OmniDocBench and better performance on tasks like table and ancient Chinese document recognition.

New

• PaddleOCR-VL-1.6: New document parsing model with improved accuracy and robustness.
• API SDKs: Official SDKs released for Python, Go, and TypeScript (software development kits) to simplify integration with PaddleOCR APIs.
• Multi-page TIFF Support: Added support for parsing multi-page TIFF images.

Fixes Worth Knowing

• Improved parsing robustness and recognition accuracy across various document types.

Before You Upgrade

Existing PaddleOCR-VL-1.5 users can upgrade to PaddleOCR-VL-1.6 with zero code changes.

TL;DR

PaddleOCR now includes a new Chinese recognition model (ch_doc) optimized for documents with a large character set, improving accuracy on complex texts.

Breaking

None.

New

• Chinese Document Recognition: Added ch_doc language model for improved accuracy on document images with over 15,000 characters.
• Environment Variable for Models: You can now specify the base directory for PaddleOCR models using the PADDLE_OCR_BASE_DIR environment variable.

Fixes Worth Knowing

• Fixed NaN issues in PP-OCRv4, improving stability.
• Resolved inference errors with KIE (Key Information Extraction) mode in ONNX format.
• Corrected issues with LaTeXOCR inference and training.
• Fixed a bug causing errors when exporting images without text to docx format.

Before You Upgrade

If you were relying on specific model download locations, consider setting the PADDLE_OCR_BASE_DIR environment variable to ensure consistent behavior.

TL;DR

PaddleOCR now offers a 3.5MB ultra-lightweight OCR (Optical Character Recognition) system, enabling deployment on a wider range of devices including mobile and embedded systems.

New

• Ultra-lightweight model: Deploy OCR on server, mobile, embedded, and IoT devices.

Fixes Worth Knowing

None.

Before You Upgrade

None.