GitHub radar
One-Shot Long Document OCR from Baidu
A new OCR tool was announced by Baidu, which enables parsing of full documents or PDFs at once.
Unlimited-OCR is a variant of Baidu’s DeepSeek-OCR, allowing you to parse multiple pages of documents/PDFs at once via a single request rather than individually. Also supports batch inference, real-time streaming inference response and cross-platform inference. Launched less than a week ago but already got nearly 400 stars.
Why a vibe-coder should care
This is a godsend if you’ve ever struggled trying to extract text from a large PDF and having to break it down into individual pages. With just one request, you’ll receive the entire text returned in a nicely structured format.
▌ More finds