1 Release published by 1 person

• v2.36.1

  published Jun 4, 2025

10 Pull requests merged by 7 people

• fix:NoneTypeError in MsPowerpointDocumentBackend

  #1747 merged Jun 10, 2025

• feat: support xlsm files

  #1520 merged Jun 10, 2025

• fix: prov for merged-elems

  #1728 merged Jun 10, 2025

• fix: initialize df_osd to avoid uninitialized variable error

  #1718 merged Jun 10, 2025

• fix: allow custom torch_dtype in vlm models

  #1735 merged Jun 10, 2025

• docs: add open webui

  #1734 merged Jun 10, 2025

• fix: Improve extraction from textboxes in Word docs

  #1701 merged Jun 6, 2025

• fix: Add WEBP to the list of image file extensions

  #1711 merged Jun 5, 2025

• fix: remove typer and click constraints

  #1707 merged Jun 4, 2025

• docs: flash-attn usage and install

  #1706 merged Jun 4, 2025

3 Pull requests opened by 3 people

• fix: #1469 handle list index out of range error with page_range param

  #1717 opened Jun 5, 2025

• fix: Handle multiple formatted elements in list MD parsing

  #1725 opened Jun 6, 2025

• feat: Make Page.parsed_page the only source of truth for text cells, add OCR cells to it

  #1745 opened Jun 10, 2025

8 Issues closed by 4 people

• pptx parsing

  #1742 closed Jun 10, 2025

• DoclingDocument validation fails with "Nothing to be deleted" error on list_item elements

  #1741 closed Jun 10, 2025

• ValueError: ListItem's parent must be a list group

  #1732 closed Jun 8, 2025

• Legitimate duplicate text in textbox in docx is being unexpectedly removed

  #1668 closed Jun 6, 2025

• Loading Failure of Files with Chinese Filenames

  #1218 closed Jun 5, 2025

• Controlled requests to external inference provider.

  #1661 closed Jun 5, 2025

• Load DoclingDocument from .doctags file?

  #1713 closed Jun 5, 2025

• How to Avoid Duplicate Table Content in Text Extraction with Docling

  #1675 closed Jun 5, 2025

21 Issues opened by 21 people

• Documents output filled with GLYPH word

  #1744 opened Jun 10, 2025

• Error while parsing : maximum recursion depth exceeded

  #1743 opened Jun 10, 2025

• Docling does not work well with Vietnamese

  #1740 opened Jun 10, 2025

• Table omission in conversion from Docling Document to markdown

  #1738 opened Jun 9, 2025

• Words in a line break into multiple lines

  #1737 opened Jun 9, 2025

• error running the simple convert example from the home page

  #1736 opened Jun 9, 2025

• Add LangChain ChainVlmOptions to complete the conversion of Vlm documents with the help of langchian

  #1733 opened Jun 8, 2025

• Use flash attention with VLM

  #1730 opened Jun 7, 2025

• Improce VLM API model calls

  #1729 opened Jun 7, 2025

• New RapidOCR Version Support

  #1727 opened Jun 6, 2025

• No CLI or docling-serve option to set truncation for tokenizer; warning cannot be suppressed

  #1726 opened Jun 6, 2025

• KeyError: -1

  #1722 opened Jun 5, 2025

• [Bee] Add OCR-derived word cells into SegmentedPage

  #1721 opened Jun 5, 2025

• Layout Parser Fine-Tune?

  #1719 opened Jun 5, 2025

• Error while parsing a .csv file

  #1716 opened Jun 5, 2025

• Conversion seems slow, while GPU is underutilized - num_threads seem to have absolutely no effect

  #1715 opened Jun 5, 2025

• It says "Input pdf "filename.pdf" is not valid, when ran as an .exe

  #1714 opened Jun 5, 2025

• Severe Markdown Conversion Issues: Cropped/Missing Images, Mixed Layout

  #1712 opened Jun 5, 2025

• CLI exporting of images from Excel and PowerPoint files fails

  #1710 opened Jun 4, 2025

• Add Comparison Chart to Top 8 Alternatives

  #1709 opened Jun 4, 2025

• Bug: Miss first list item when new list begins

  #1705 opened Jun 4, 2025

29 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• fix: pptx line break and space handling

  #1664 commented on Jun 10, 2025 • 4 new comments

• fix: pptx shape order

  #1658 commented on Jun 10, 2025 • 0 new comments

• fix(msword_backend): Identify text in the same line after an image #1425

  #1610 commented on Jun 10, 2025 • 0 new comments

• fix: Capturing of pptx images following the docx backend

  #1328 commented on Jun 7, 2025 • 0 new comments

• Getting word-level bounding boxes from DocumentConverter

  #1704 commented on Jun 10, 2025 • 0 new comments

• Performance Issue. 28.1% less inference time on demo case with a simple change.

  #1521 commented on Jun 10, 2025 • 0 new comments

• Poor performance on pages with many elements

  #1624 commented on Jun 10, 2025 • 0 new comments

• How do I add GPU parallelization for the do_formula_enrichment model?

  #1693 commented on Jun 10, 2025 • 0 new comments

• Inconsistent Markdown Output with generate_multimodal_pages Method in Docling

  #1584 commented on Jun 10, 2025 • 0 new comments

• Bugs new Docling release

  #1703 commented on Jun 9, 2025 • 0 new comments

• Embedded Image data was not loaded correctly when opening a MD file

  #1305 commented on Jun 9, 2025 • 0 new comments

• CUDA support

  #1649 commented on Jun 9, 2025 • 0 new comments

• Issue: EasyOCR model cannot be accessed from Spark workers when using pre-downloaded docling model

  #1414 commented on Jun 8, 2025 • 0 new comments

• Refined layout parsing

  #1614 commented on Jun 8, 2025 • 0 new comments

• Downloading detection model, please wait. This may take several minutes depending upon your network connection.

  #1638 commented on Jun 7, 2025 • 0 new comments

• Export to markdown does not save image references

  #1574 commented on Jun 6, 2025 • 0 new comments

• Footer Text Interferes with Main PDF Content During Parsing

  #1625 commented on Jun 6, 2025 • 0 new comments

• Stride - chunk overlap

  #1686 commented on Jun 5, 2025 • 0 new comments

• As multiple PDFs are parsed, the memory of Docling continues to increase and will not decrease.

  #1311 commented on Jun 5, 2025 • 0 new comments

• Incorrectly populated ProvenanceItem in ReadingOrderModel

  #1699 commented on Jun 5, 2025 • 0 new comments

• Runtime Error since v.2.34.0 related to OSD detection

  #1657 commented on Jun 5, 2025 • 0 new comments

• Memory leak caused by EasyOCR

  #1343 commented on Jun 5, 2025 • 0 new comments

• Multi-threading and multi-processing for faster parsing

  #1256 commented on Jun 5, 2025 • 0 new comments

• Incorrect Table Columns

  #1678 commented on Jun 5, 2025 • 0 new comments

• Export to markdown only contains H2 headers

  #1023 commented on Jun 4, 2025 • 0 new comments

• html parser KeyError: -1

  #1702 commented on Jun 4, 2025 • 0 new comments

• Coversion of the document contains false positive classified tables

  #1680 commented on Jun 4, 2025 • 0 new comments

• KeyError and RuntimeError occurred when opening a document(docx)

  #1645 commented on Jun 4, 2025 • 0 new comments

• Docling getting killed when i feed bigger pdf files which have 900+ pages

  #1654 commented on Jun 4, 2025 • 0 new comments