surya_layout is a vision-encoder-decoder model developed by Datalab and published on Hugging Face. It is designed for document layout analysis, identifying structural elements such as titles, paragraphs, tables, and images within PDFs and scanned pages. The model is lightweight (0.1B parameters) and distributed under a CC-BY-NC-SA license. It is commonly used as a component in document processing pipelines, including OCR and PDF parsing tools. The model is intended for developers integrating layout understanding into document intelligence applications.
Surya Layout sits in PulseGate's Other AI category. Accurately detecting and classifying layout elements in scanned or digital documents for downstream OCR and parsing tasks. It is built as an open-source project for developers building document AI systems. The project is open source (Open Source). It ships for the web, and it can be self-hosted.
Datalab builds and maintains Surya Layout, and it first shipped in 2024. Among its 3 catalogued features are Layout Analysis, Document Parsing, and Vision Encoder.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do