surya
OCR, layout analysis, reading order, table recognition in 90+ languages
File Explorer
Download Latest Version (.zip)- breaking-bug-report.md
- feature_request.md
- output-bug-report.md
- ci.yml
- cla.yml
- publish.yml
- scripts.yml
- cla.json
- multilingual.md
- .gitignore
- corporate.png
- corporate_layout.png
- corporate_reading.png
- corporate_tablerec.png
- corporate_text.png
- excerpt.png
- excerpt_layout.png
- excerpt_reading.png
- excerpt_text.png
- form.png
- form_layout.png
- form_reading.png
- form_tablerec.png
- form_text.png
- handwritten.png
- handwritten_layout.png
- handwritten_reading.png
- handwritten_tablerec.png
- handwritten_text.png
- latex_ocr.png
- newspaper.png
- newspaper_layout.png
- newspaper_reading.png
- newspaper_text.png
- olmocr_size_chart.png
- scanned_tablerec.png
- textbook.png
- textbook_layout.png
- textbook_reading.png
- textbook_text.png
- datalab-logo.png
- __init__.py
- client.py
- config.py
- serialize.py
- server.py
- __init__.py
- order_ar.py
- predictor.py
- dinov2_base.json
- dinov2_large.json
- dinov2_small.json
- dinov2_with_registers_base.json
- dinov2_with_registers_large.json
- dinov2_with_registers_small.json
- __init__.py
- backbone.py
- base.py
- dinov2.py
- dinov2_with_windowed_attn.py
- projector.py
- __init__.py
- ms_deform_attn_func.py
- __init__.py
- ms_deform_attn.py
- __init__.py
- __init__.py
- lwdetr.py
- position_encoding.py
- transformer.py
- __init__.py
- box_ops.py
- logger.py
- misc.py
- __init__.py
- predictor.py
- __init__.py
- blank.py
- compat.py
- load.py
- polygon.py
- predictor.py
- pretrained.py
- rfdetr_torch.py
- s3.py
- util.py
- draw.py
- fonts.py
- katex.js
- render_html.py
- text.py
- __init__.py
- config.py
- encoderdecoder.py
- __init__.py
- client.py
- config.py
- heatmap.py
- loader.py
- parallel.py
- processor.py
- schema.py
- server.py
- util.py
- __init__.py
- client.py
- config.py
- server.py
- __init__.py
- base.py
- llamacpp.py
- openai_client.py
- spawn.py
- vllm.py
- __init__.py
- parsers.py
- prompts.py
- schema.py
- util.py
- load.py
- processing.py
- __init__.py
- label.py
- schema.py
- __init__.py
- config.py
- __init__.py
- client.py
- config.py
- loader.py
- schema.py
- server.py
- __init__.py
- schema.py
- surya_screenshot.html
- __init__.py
- build_gguf.py
- config.py
- detect_layout.py
- detect_text.py
- ocr_text.py
- run_streamlit_app.py
- screenshot_app.py
- streamlit_app.py
- table_recognition.py
- __init__.py
- schema.py
- __init__.py
- logging.py
- settings.py
- test_latex.png
- conftest.py
- test_detection.py
- test_fast_layout_server.py
- test_layout.py
- test_ocr_errors.py
- test_recognition.py
- test_table_rec.py
- .gitignore
- .pre-commit-config.yaml
- CITATION.cff
- CLA.md
- LICENSE
- MODEL_LICENSE
- pyproject.toml
- pytest.ini
- README.md
- uv.lock
# Installation Guide
1. Get the code
git clone https://github.com/datalab-to/surya
Downloads the entire project code from GitHub to your computer.
cd surya
Moves into the project folder you just downloaded.
2. Official Install Script
Easy Recommendedpip install surya-ocr
Installs the package published on PyPI directly β no need to clone the source.
brew install llama.cpp # macOS
Installs the pre-built package via Homebrew β no source build required.
pip install streamlit pdftext
Installs the package published on PyPI directly β no need to clone the source.
After installing, open a new terminal and run the program's version command (e.g. --version) to confirm it worked.
Pulled directly from this repo's README.
3. Python
EasyPrerequisites
pip install surya-ocr
Installs the package published on PyPI directly β no need to clone the source.
pip install streamlit pdftext
Installs the package published on PyPI directly β no need to clone the source.
If it runs without errors and prints output in the terminal, it worked.
Pulled directly from this repo's README.
// repository documentation
Was this content helpful?
(0 ratings)