Commit Graph

  • 7a4593f41f chore: bump version to 2.66.0 [skip ci] main v2.66.0 github-actions[bot] 2025-12-24 10:46:23 +00:00
  • 183900d5b3 Deployed faff935 with MkDocs version: 1.6.1 gh-pages 2025-12-24 10:15:52 +00:00
  • faff935b0e
    fix(docx): handle tables with merged cells causing IndexError (#2813) madmaxxt 2025-12-24 12:13:07 +02:00
  • be085c0e39
    docs(RTX): Guidelines for best performance on RTX GPUs (#2765) Michele Dolfi 2025-12-19 13:16:59 +01:00
  • cc5e3cee74
    docs: add docstrings to DocumentConverter #2748 (#2782) Julia Pap 2025-12-19 11:20:33 +01:00
  • 150fe90728
    docs(style): fix link visibility in dark mode (#2804) Shivaditya Meduri 2025-12-18 16:08:11 +01:00
  • f21ae74016 update result with initial text feat-deepseek-ocr Michele Dolfi 2025-12-17 14:03:37 +01:00
  • b9647dcf38 Merge remote-tracking branch 'origin/main' into feat-deepseek-ocr Michele Dolfi 2025-12-17 14:02:43 +01:00
  • 595115d892
    fix(markdown): allow text before headers also in mixed markdown and html (#2801) Michele Dolfi 2025-12-17 13:54:07 +01:00
  • c3204d034f fix broken html in test Michele Dolfi 2025-12-17 11:00:21 +01:00
  • aaab28f0ab add parsing of annotated markdown and definition of new ResponseFormat for the VLM pipeline Michele Dolfi 2025-12-17 10:10:20 +01:00
  • 241d19ed6f
    feat: add preset for using granite-docling via vllm and other apis (#2792) Michele Dolfi 2025-12-16 17:22:35 +01:00
  • 799f284f82 chore: bump version to 2.65.0 [skip ci] v2.65.0 github-actions[bot] 2025-12-15 16:54:58 +00:00
  • 292b5053cf refactor(asr): use ProvenanceTrack in ASR pipeline dev/webvtt-refactor Cesar Berrospi Ramis 2025-12-15 00:30:47 +01:00
  • e2e09a0638 style(asr): remove unnecessary imports Cesar Berrospi Ramis 2025-12-15 00:04:11 +01:00
  • 1212f3b9fb refactor(webvtt): set ProvenanceTrack timinings as float type Cesar Berrospi Ramis 2025-12-14 23:59:42 +01:00
  • 3361c6b76d refactor(webvtt): preserve new lines and add helper handlers Cesar Berrospi Ramis 2025-12-14 11:22:10 +01:00
  • 95813eb909 refactor(webvtt): update WebVTTDocumentBackend with new docling-core classes Cesar Berrospi Ramis 2025-12-12 13:25:52 +01:00
  • 9788a0b761 refactor(provenance): account for provenance as union of ProvenanceItem and ProvenanceTrack Cesar Berrospi Ramis 2025-12-08 17:10:50 +01:00
  • 4897092b0b
    test: update verify_utils to check CodeItem and FormulaItem (#2775) Cesar Berrospi Ramis 2025-12-12 13:10:33 +01:00
  • 7c24b014f6
    docs: add Pydantic field documentation for PipelineOptions (#2771) Nikolaos Georgantopoulos 2025-12-11 13:30:41 +01:00
  • 807303e33e
    chore: mkdocstring python handler to render pydantic field (#2770) Edoardo Abati 2025-12-11 12:54:13 +01:00
  • d03439ccc5
    docs(gpu): Add benchmarks of standard pipeline with OCR (#2764) Michele Dolfi 2025-12-10 20:43:20 +01:00
  • da7678a754
    feat: Add YAML output format to CLI (#2768) Nick Hoernle 2025-12-10 20:35:24 +01:00
  • 1d78418cef
    fix(rapidocr): use correct parameter name for rec_keys_path (#2762) Francesco 2025-12-10 16:44:10 +01:00
  • a97d950d74
    fix(docx): handle missing value in paragraph style name (#2761) Cesar Berrospi Ramis 2025-12-09 12:36:09 +01:00
  • 6afd7c57ff chore: bump version to 2.64.1 [skip ci] v2.64.1 github-actions[bot] 2025-12-09 09:06:09 +00:00
  • 1df0560ec2
    fix: Clear word/char cells when force_full_page_ocr is used (#2738) Myles 2025-12-09 03:33:09 -05:00
  • edbabfcac2
    fix: add missing font download in the rapidocr artifacts (#2735) Michele Dolfi 2025-12-08 12:44:53 +01:00
  • 609069d12c
    fix: Ensure proper image_scale for generated page images in VLM pipelines (#2728) Christoph Auer 2025-12-05 13:16:11 +01:00
  • d007ba0e6f
    fix(html): tackle paragraphs with block-level elements (#2720) Cesar Berrospi Ramis 2025-12-05 12:52:53 +01:00
  • 3df3cf8664
    fix: add page as argument to build_prompt elh/update_2stage_inference ElHachem02 2025-12-04 13:36:20 +01:00
  • aebe25cf00
    fix(html): prevent hierarchy reset in rich table cells (#2716) Matvei Smirnov 2025-12-03 20:52:23 +03:00
  • 0904dbb95a
    feat: update inference code to shuffle layout elements and discard initial prompt ElHachem02 2025-12-03 12:59:31 +01:00
  • 92e4f2220a Fix artifacts_path handling in Layout+VLM pipeline cau/fix-layout-vlm-pipeline-artifacts-path Christoph Auer 2025-12-03 12:52:22 +01:00
  • c97715f5fd
    fix(docx): parse integrals as n-ary objects without chr element (#2712) Cesar Berrospi Ramis 2025-12-03 11:25:52 +01:00
  • f80c903c24 chore: bump version to 2.64.0 [skip ci] v2.64.0 github-actions[bot] 2025-12-02 11:25:22 +00:00
  • 6ef4ffd643
    fix: InputFormat.IMAGE must have correct pipeline (#2707) Christoph Auer 2025-12-01 19:44:16 +01:00
  • 5bbc94daf8 Add page image injection cau/layout-vlm-pipeline-page-images Christoph Auer 2025-12-01 15:20:41 +01:00
  • 54cd6d7406
    fix: do not consider singleton cells in xlsx as TableItems but rather TextItems (#2589) glypt 2025-11-27 16:25:32 +01:00
  • c0b57ae389
    chore: Cleaning the example of post_process_ocr_with_vlm (#2693) Maxim Lysak 2025-11-27 12:38:45 +01:00
  • fa21128138
    docs: Example on how to apply external OCR as post processing (#2517) Maxim Lysak 2025-11-27 11:04:40 +01:00
  • 0049857c7d
    chore: update mlx lock (#2689) Panos Vagenas 2025-11-27 10:25:07 +01:00
  • 134436245a
    feat(experimental): Add experimental TableCropsLayoutModel (#2669) Christoph Auer 2025-11-25 05:14:51 +01:00
  • b75c6461f4
    docs: More GPU results and improvements in the example docs (#2674) Michele Dolfi 2025-11-24 15:26:08 +01:00
  • 146b4f0535
    docs: fix typo on jobkit page (#2671) Muhammad Ali Hasan 2025-11-24 02:35:45 -06:00
  • e58055465c
    fix(docx): Missing list items after numbered header (#2665) Michele Dolfi 2025-11-24 08:49:21 +01:00
  • ad97e52851
    feat: Factory and plugin-capability for Layout and Table models (#2637) Christoph Auer 2025-11-21 10:26:06 +01:00
  • dcb57bf528 chore: bump version to 2.63.0 [skip ci] v2.63.0 github-actions[bot] 2025-11-20 14:42:37 +00:00
  • 2087c6bf9f
    fix: Respect document_timeout in new threaded StandardPdfPipeline (#2653) Christoph Auer 2025-11-20 14:57:14 +01:00
  • 54e65d9511
    chore: update Milvus on examples and references to deprecated method (#2664) Cesar Berrospi Ramis 2025-11-20 13:22:45 +01:00
  • ce5a099dfd
    docs: Add Hector as compatible AI agent platform integration (#2662) kadirpekel 2025-11-20 13:02:47 +01:00
  • b559813b9b
    feat: add save and load for conversion result (#2648) Peter W. J. Staar 2025-11-20 12:45:26 +01:00
  • 6fb9a5f98a
    fix: In DocumentConverter.convert_string() make nullable name parameter optional (#2660) Cristi Burcă 2025-11-20 05:24:27 +00:00
  • 463a3fd474
    fix: Enable GPU for RapidOCR when available (#2659) Michele Dolfi 2025-11-19 17:12:00 +01:00
  • b216ad848d
    docs: Added documentation to use SuryaOCR via plugin docling-surya (#2533) Harry Ho 2025-11-19 22:27:24 +08:00
  • 6fe6aae91a Apply ruff formatting to test file copilot/fix-page-range-bug copilot-swe-agent[bot] 2025-11-19 13:28:01 +00:00
  • 0788e714a9 Add comprehensive tests for page_range bug fix copilot-swe-agent[bot] 2025-11-19 13:26:25 +00:00
  • 58fc6ccf86 Fix page_range stopping at page 32 by using dynamic batch_size copilot-swe-agent[bot] 2025-11-19 13:25:00 +00:00
  • 18f705b235 Initial plan copilot-swe-agent[bot] 2025-11-19 13:10:27 +00:00
  • 03e7c7d924
    docs: Fix broken homepage links (#2651) Robyn Johnson 2025-11-19 01:19:56 -06:00
  • 8af228f1e2
    docs(examples): processing parquet file of images (#2641) Michele Dolfi 2025-11-19 06:39:25 +01:00
  • da4c2e9dbe
    fix: remove py3.14 requirement for default rapidocr (#2639) Michele Dolfi 2025-11-18 17:23:43 +01:00
  • d549445e78
    docs: Move Installation and Quickstart (Usage) under Getting started (#2644) Ryan Soliveres 2025-11-19 00:09:41 +08:00
  • ac9fc585bb
    docs: add redirection from getting started page (#2640) Panos Vagenas 2025-11-17 14:13:51 +01:00
  • f5528623a7
    docs(examples): remove deprecation warnings with export_to_dataframe (#2638) Cesar Berrospi Ramis 2025-11-17 12:48:41 +01:00
  • d6ddf9f4cb chore: bump version to 2.62.0 [skip ci] v2.62.0 github-actions[bot] 2025-11-17 11:34:08 +00:00
  • 3495b73de8
    feat: add the Image backend (#2627) Peter W. J. Staar 2025-11-17 11:37:22 +01:00
  • aa75dd13d3 test: mark timeout test as manual due to model requirement copilot/fix-document-timeout-bug copilot-swe-agent[bot] 2025-11-17 09:27:27 +00:00
  • e3aa8cd770 feat: add document_timeout support to StandardPdfPipeline copilot-swe-agent[bot] 2025-11-17 09:23:28 +00:00
  • f3ed123b51 Initial plan copilot-swe-agent[bot] 2025-11-17 09:17:41 +00:00
  • ae30373ee7
    docs: combine Home and Getting Started pages (#2600) Robyn Johnson 2025-11-14 06:29:25 -06:00
  • 14b436d590
    fix: correct the model-repo name (#2624) Peter W. J. Staar 2025-11-14 13:21:08 +01:00
  • 55908d6bb4 chore: pretest docling-core 2.51.0 pretest-core-2-51-0 Panos Vagenas 2025-11-12 16:35:49 +01:00
  • bbb66d8be0 Add documentation for reading order patch copilot/fix-keyerror-in-docling copilot-swe-agent[bot] 2025-11-12 13:07:43 +00:00
  • 570fe949c9 Add monkey patch to fix KeyError in reading order model copilot-swe-agent[bot] 2025-11-12 13:03:50 +00:00
  • 609988d3e1 Initial plan copilot-swe-agent[bot] 2025-11-12 12:48:22 +00:00
  • 4852d8b4f2
    feat(experimental): Layout + VLM model with layout prompt (#2244) Christoph Auer 2025-11-12 13:42:09 +01:00
  • 054c4a634d
    fix(docx): parse page headers and footers (#2599) Cesar Berrospi Ramis 2025-11-10 16:10:12 +01:00
  • 463051b852 chore: bump version to 2.61.2 [skip ci] v2.61.2 github-actions[bot] 2025-11-10 11:44:59 +00:00
  • 5c27567c41
    fix: default to EasyOCR in Python 3.14 (#2605) Panos Vagenas 2025-11-10 12:09:00 +01:00
  • 06ae8ae29a
    chore: replace ds4sd with docling-project (#2596) Peter W. J. Staar 2025-11-07 11:25:56 +01:00
  • c21327cd74 chore: bump version to 2.61.1 [skip ci] v2.61.1 github-actions[bot] 2025-11-06 05:19:20 +00:00
  • ef623ffcee
    fix(docx): slow table parsing (#2553) Cesar Berrospi Ramis 2025-11-06 05:25:53 +01:00
  • 0ba8d5d9e3
    fix(html): slow table parsing (#2582) Cesar Berrospi Ramis 2025-11-06 05:25:36 +01:00
  • 8da3d287ed
    docs: make navigation menus collapse and expand (#2573) Robyn Johnson 2025-11-05 22:25:19 -06:00
  • 0ccc0a3245 chore: bump version to 2.61.0 [skip ci] v2.61.0 github-actions[bot] 2025-11-06 04:25:06 +00:00
  • fa925741b6
    fix: temporarily pin NuExtract to working revision (#2588) Panos Vagenas 2025-11-05 21:23:12 +01:00
  • 8940045463 replace match with if docs/add-extraction-script Peter Staar 2025-11-05 16:57:16 +01:00
  • 1ec6c58b95 adding extraction script Peter Staar 2025-11-05 15:43:56 +01:00
  • 6a04e27352
    feat(vlm): track generated tokens and stop reasons for VLM models (#2543) peets 2025-11-04 19:39:09 +01:00
  • 1a5146abc9
    fix(ocr): use PSM integer values directly instead of constructor (#2578) 정물결 2025-11-05 03:32:41 +09:00
  • 32a5aed5ea chore: bump version to 2.60.1 [skip ci] v2.60.1 github-actions[bot] 2025-11-04 11:26:12 +00:00
  • 0e1b0bd816
    chore: switch print statements to debug logging (#2569) Panos Vagenas 2025-11-04 11:32:39 +01:00
  • fb737d026e
    chore: fix malformed f-string (#2563) Johannes Damp 2025-11-04 11:01:26 +01:00
  • 8360aa5449
    fix: extract response from api_image_request in picture description (#2571) peets 2025-11-04 08:39:15 +01:00
  • 3467b0a035 chore: bump version to 2.60.0 [skip ci] v2.60.0 github-actions[bot] 2025-10-31 14:43:29 +00:00
  • 268d027c8f
    feat: Use threading in the standard pipeline and move old behavior to legacy (#2452) Michele Dolfi 2025-10-31 14:42:11 +01:00
  • 01577e92d1
    docs: Update link to Open WebUI docs (#2549) Welteam 2025-10-31 12:21:11 +00:00
  • cb100437fa
    docs: Update installation options with extras and review FAQ (#2548) Michele Dolfi 2025-10-31 13:21:01 +01:00