[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"readme:marker":3},"\u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fdatalab-to\u002Fmarker\u002FHEAD\u002Fdata\u002Fimages\u002Fdatalab-logo.png\" alt=\"Datalab Logo\" width=\"150\" \u002F>\n\u003C\u002Fp>\n\u003Ch1>Datalab\u003C\u002Fh1>\n\u003Cp align=\"center\">\n  \u003Cstrong>State of the Art models for Document Intelligence\u003C\u002Fstrong>\n\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode%20License-Apache%202.0-green.svg\" alt=\"Code License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.datalab.to\u002Fpricing\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel%20License-OpenRAIL--M-blue.svg\" alt=\"Model License\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002FKuZwXNGnfH\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-Join%20us-5865F2?logo=discord&logoColor=white\" alt=\"Discord\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.datalab.to\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHomepage-datalab.to-blue\" alt=\"Homepage\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fdocumentation.datalab.to\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocs-Read%20the%20docs-blue\" alt=\"Docs\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.datalab.to\u002Fplayground\" rel=\"nofollow ugc noopener\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPlayground-Try%20it-orange\" alt=\"Public Playground\" \u002F>\u003C\u002Fa>\n\u003C\u002Fp>\u003Chr \u002F>\u003Ch1>Marker\u003C\u002Fh1>\n\u003Cp>Marker converts documents to markdown, JSON, chunks, and HTML quickly and accurately.\u003C\u002Fp>\n\u003Cul>\n\u003Cli>Converts PDF, image, PPTX, DOCX, XLSX, HTML, EPUB files in all languages\u003C\u002Fli>\n\u003Cli>Formats tables, forms, equations, inline math, links, references, and code blocks\u003C\u002Fli>\n\u003Cli>Extracts and saves images\u003C\u002Fli>\n\u003Cli>Removes headers\u002Ffooters\u002Fother artifacts\u003C\u002Fli>\n\u003Cli>Extensible with your own formatting and logic\u003C\u002Fli>\n\u003Cli>Optionally boost accuracy with LLMs (and your own prompt)\u003C\u002Fli>\n\u003Cli>Works on GPU, CPU, or MPS\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2>Try Datalab's Managed Platform\u003C\u002Fh2>\n\u003Cp>Our managed platform runs a version of our latest open source model, \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fdatalab-to\u002Fchandra\" rel=\"nofollow ugc noopener\">Chandra\u003C\u002Fa> — higher accuracy than Marker, with zero data retention by default, SOC 2 Type 2, and custom BAAs.\u003C\u002Fp>\n\u003Cp>If you have high volume workloads, we offer a batch processing service that has processed 1B+ pages per week — we manage the infrastructure so your workloads finish on time.\u003C\u002Fp>\n\u003Cp>Get started with \u003Cstrong>$5 in free credits\u003C\u002Fstrong> — \u003Ca href=\"https:\u002F\u002Fwww.datalab.to\u002F?utm_source=gh-marker\" rel=\"nofollow ugc noopener\">sign up\u003C\u002Fa>.\u003C\u002Fp>\n\u003Ch2>Performance\u003C\u002Fh2>\n\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fdatalab-to\u002Fmarker\u002FHEAD\u002Fdata\u002Fimages\u002Folmocr_bench.png\" width=\"800px\" \u002F>\u003Cp>We measure marker on \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fallenai\u002Folmocr\u002Ftree\u002Fmain\u002Folmocr\u002Fbench\" rel=\"nofollow ugc noopener\">olmocr-bench\u003C\u002Fa>, a third-party benchmark of 1,403 PDFs with tests covering math, tables, multi-column layout, scans, and hard edge cases.  Balanced mode scores \u003Cstrong>76.0%\u003C\u002Fstrong> overall — \u003Cstrong>83.5%\u003C\u002Fstrong> on born-digital PDFs — ahead of MinerU and docling and within range of much larger VLMs, while fast mode runs the layout + text-layer path far cheaper (and a no-OCR mode goes faster still).  Scores are the olmocr-bench overall (macro-average across the 8 categories).\u003C\u002Fp>\n\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fdatalab-to\u002Fmarker\u002FHEAD\u002Fdata\u002Fimages\u002Folmocr_digital.png\" width=\"700px\" \u002F>\u003Cp>See \u003Ca href=\"#benchmarks\" rel=\"nofollow ugc noopener\">below\u003C\u002Fa> for the full per-category scores, the competitive comparison, and instructions on how to run your own benchmarks.\u003C\u002Fp>\n\u003Ch2>Hybrid Mode\u003C\u002Fh2>\n\u003Cp>For the highest accuracy, pass the \u003Ccode>--use_llm\u003C\u002Fcode> flag to use an LLM alongside marker.  This will do things like merge tables across pages, handle inline math, format tables properly, and extract values from forms.  It works with Gemini, Claude, OpenAI-compatible, Azure, Vertex, OpenRouter, or Ollama models.  By default, it uses \u003Ccode>gemini-3.5-flash\u003C\u002Fcode>.  See \u003Ca href=\"#llm-services\" rel=\"nofollow ugc noopener\">below\u003C\u002Fa> for details.\u003C\u002Fp>\n\u003Ch2>Examples\u003C\u002Fh2>\n\u003Ctable>\n\u003Cthead>\n\u003Ctr>\n\u003Cth>PDF\u003C\u002Fth>\n\u003Cth>File type\u003C\u002Fth>\n\u003Cth>Markdown\u003C\u002Fth>\n\u003Cth>JSON\u003C\u002Fth>\n\u003C\u002Ftr>\n\u003C\u002Fthead>\n\u003Ctbody>\u003Ctr>\n\u003Ctd>\u003Ca href=\"https:\u002F\u002Fgreenteapress.com\u002Fthinkpython\u002Fthinkpython.pdf\" rel=\"nofollow ugc noopener\">Think Python\u003C\u002Fa>\u003C\u002Ftd>\n\u003Ctd>Textbook\u003C\u002Ftd>\n\u003Ctd>[View](\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FVikParuchuri\u002Fmarker\u002Fblob\u002Fmaster\u002Fdata\u002F\" rel=\"nofollow ugc noopener\">https:\u002F\u002Fgithub.com\u002FVikParuchuri\u002Fmarker\u002Fblob\u002Fmaster\u002Fdata\u002F\u003C\u002Fa>\u003C\u002Ftd>\n\u003Ctd>\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftbody>\u003C\u002Ftable>\n",1787581367032]