Heads up: The Hugging Face demo can be slow at times. For a faster experience, please try Baidu AI Studio or ModelScope.
( Use this mode for recognizing full-page documents with structured layouts, such as reports, papers, or magazines.)
💡 To recognize a single, pre-cropped element (e.g., a table or formula), switch to the 'Element-level Recognition' tab for better results.
History Updates
-
Oct 30, 2025:
Added two advanced control options under the "Document Parsing" tab. These features were previously enabled by default (set to true) but are now user-configurable and default to false.
- Enable document unwarping: Corrects distortions in bent or poorly photographed documents.
- Enable orientation classification: Automatically corrects the orientation of rotated or upside-down images.
- Oct 16, 2025: Initial release of the demo.
Please upload an image and click 'Parse Document'.
(Best for images with a simple, single-column layout (e.g., pure text), or for a pre-cropped single element like a table, formula, or chart.)
Choose a recognition type:
Please upload an image and click a recognition type.