Baidu has released and open-sourced its end-to-end OCR model, Unlimited OCR, and it ranked first on multiple GitHub and Hugging Face trend lists the next day.
According to Jin10, Unlimited OCR topped GitHub’s Daily Trending list and the Python list, and also ranked No. 1 on Hugging Face’s global overall model trending list and multimodal model trending list.
Baidu said the model was built for long-document parsing. It has 3 billion total parameters, with about 570 million activated parameters during inference.
Public evaluation results showed Unlimited OCR achieved a 93.92% overall score on the OmniDocBench v1.6 benchmark test, setting a new record for end-to-end OCR performance.
AI TRENDS | Baidu Open-Sources Unlimited OCR Model, Tops GitHub and Hugging Face Trend Lists
2026-06-29 14:15:42
Disclaimer:
1. The information provided does not constitute investment advice. Investors should make independent decisions and bear all risks themselves.
2. The copyright of this content belongs to the original author. The views expressed herein are solely those of the author and do not represent the stance or position of this website.
Previous article：
AI趋势 | 百度Unlimited OCR开源登顶四榜
