PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters
Which summary reads better? Pick one — models revealed after.Both summaries are AI-generated.
PaddleOCR has released PP-OCRv6, its latest digital text detection and recognition model family now available on Hugging Face. The model family spans three tiers ranging from 1.5 million to 34.5 million parameters, supporting 50 languages across various document and real-world image formats. Featuring architectural upgrades like the PPLCNetV4 backbone and RepLKFPN detection module, PP-OCRv6 delivers significant accuracy improvements over its predecessor while maintaining a lightweight footprint for easy deployment.
PP-OCRv6 is the latest generation of PaddleOCR's universal OCR model, supporting 50 languages and scaling from 1.5M to 34.5M parameters across three model tiers. It achieves 86.2% detection Hmean and 83.2% recognition accuracy on PaddleOCR's in-house multi-scenario OCR benchmarks. The model is designed for real-world text detection and recognition across various documents and images.