PP-OCRv6: 1.5M-34.5M parameter OCR models surpass billion-scale VLMs

Hugging Face Models

PP-OCRv6 scales from 1.5M to 34.5M parameters, outperforming billion-parameter VLMs on OCR tasks. It achieves +4.9% detection and +5.1% recognition accuracy improvements, with models optimized for browser, edge, and server deployment.

OCRpp-ocr

See 2 more sources

PP-OCRv6: From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks15 days agoYubo Zhang, Xueqing Wang, Manhui Lin, Yue Zhang, Penglongyi Deng, Ting Sun, Tingquan Gao, Zelun Zhang, Jiaxuan Liu, Changda Zhou, Hongen Liu, Suyin Liang, Cheng Cui, Yi Liu, Dianhai Yu, Yanjun Ma

🚀PP-OCRv6 is officially released !14 days agoKokaOPDiscuss

Hugging Face Models·PaddlePaddle

Jun 9, 4:03 AM