opendatalab/DocLayout-YOLO

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2258 stars · 171 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

DocLayout-YOLO is a real-time YOLO-v10-based model for detecting document layout elements (text blocks, tables, figures, etc.) in diverse documents, trained with a large synthetic dataset (DocSynth-300K). It ships as a pip-installable Python package with pretrained weights, inference scripts, and a dataset synthesis pipeline.

Use cases

• detect layout of pdf documents
• extract tables and figures from scanned documents
• preprocess documents for ocr pipelines
• segment document pages into structural elements
• build a pdf-to-markdown extraction pipeline
• generate synthetic document layout training data

When to choose

• you need fast, real-time document layout detection with pretrained models
• you want a pip-installable library integrated into document extraction workflows like PDF-Extract-Kit
• you need a pipeline to synthesize large-scale document layout datasets

When to avoid

• you need full OCR text recognition rather than layout detection
• your project requires a permissive license since this is AGPL-3.0
• you need general-purpose object detection unrelated to documents

Facets

library · maturity active

computer-vision machine-learning deep-learning ocr pdf image-processing computer-vision pdf machine-learning artificial-intelligence python cross-platform document-layout-analysis yolo object-detection pytorch pdf-parsing pretrained-models gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="opendatalab/DocLayout-YOLO")