{"id":31462106,"url":"https://github.com/alibaba/Logics-Parsing","last_synced_at":"2025-10-01T16:01:51.824Z","repository":{"id":316003727,"uuid":"1054563053","full_name":"alibaba/Logics-Parsing","owner":"alibaba","description":null,"archived":false,"fork":false,"pushed_at":"2025-09-30T06:59:08.000Z","size":35538,"stargazers_count":266,"open_issues_count":10,"forks_count":18,"subscribers_count":3,"default_branch":"main","last_synced_at":"2025-09-30T08:41:00.054Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/alibaba.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-09-11T02:57:24.000Z","updated_at":"2025-09-30T08:30:11.000Z","dependencies_parsed_at":"2025-09-22T06:36:46.674Z","dependency_job_id":null,"html_url":"https://github.com/alibaba/Logics-Parsing","commit_stats":null,"previous_names":["alibaba/logics-parsing"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/alibaba/Logics-Parsing","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alibaba%2FLogics-Parsing","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alibaba%2FLogics-Parsing/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alibaba%2FLogics-Parsing/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alibaba%2FLogics-Parsing/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/alibaba","download_url":"https://codeload.github.com/alibaba/Logics-Parsing/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alibaba%2FLogics-Parsing/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":277866772,"owners_count":25891280,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-01T02:00:09.286Z","response_time":88,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-10-01T16:00:29.971Z","updated_at":"2025-10-01T16:01:51.818Z","avatar_url":"https://github.com/alibaba.png","language":"Python","funding_links":[],"categories":["数据 Data","光学字符识别OCR"],"sub_categories":["资源传输下载"],"readme":"\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"imgs/logo.jpg\" width=\"80%\" \u003e\n\u003c/div\u003e\n\n\n\u003cp align=\"center\"\u003e\n    🤗 \u003ca href=\"https://huggingface.co/Logics-MLLM/Logics-Parsing\"\u003eModel\u003c/a\u003e\u0026nbsp\u0026nbsp | \u0026nbsp\u0026nbsp🤖 \u003ca href=\"https://www.modelscope.cn/studios/Alibaba-DT/Logics-Parsing/summary\"\u003eDemo\u003c/a\u003e\u0026nbsp\u0026nbsp | \u0026nbsp\u0026nbsp📑 \u003ca href=\"https://arxiv.org/abs/2509.19760\"\u003eTechnical Report\u003c/a\u003e\n\u003c/p\u003e\n\n## Introduction\n\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"imgs/overview.png\" alt=\"LogicsDocBench 概览\" style=\"width: 800px; height: 250px;\"\u003e\n\u003c/div\u003e\n\n\u003cdiv align=\"center\"\u003e\n  \u003ctable style=\"width: 800px;\"\u003e\n    \u003ctr\u003e\n      \u003ctd align=\"center\"\u003e\n        \u003cimg src=\"imgs/report.gif\" alt=\"研报示例\"\u003e\n      \u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\n        \u003cimg src=\"imgs/chemistry.gif\" alt=\"化学分子式示例\"\u003e\n      \u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\n        \u003cimg src=\"imgs/paper.gif\" alt=\"论文示例\"\u003e\n      \u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\n        \u003cimg src=\"imgs/handwritten.gif\" alt=\"手写示例\"\u003e\n      \u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd align=\"center\"\u003e\u003cb\u003ereport\u003c/b\u003e\u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\u003cb\u003echemistry\u003c/b\u003e\u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\u003cb\u003epaper\u003c/b\u003e\u003c/td\u003e\n      \u003ctd align=\"center\"\u003e\u003cb\u003ehandwritten\u003c/b\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/table\u003e\n\u003c/div\u003e\n\n\n\nLogics-Parsing is a powerful, end-to-end document parsing model built upon a general Vision-Language Model (VLM) through Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). It excels at accurately analyzing and structuring highly complex documents.\n\n## Key Features\n\n*   **Effortless End-to-End Processing**\n    *   Our single-model architecture eliminates the need for complex, multi-stage pipelines. Deployment and inference are straightforward, going directly from a document image to structured output.\n    *   It demonstrates exceptional performance on documents with challenging layouts.\n\n*   **Advanced Content Recognition**\n    *   It accurately recognizes and structures difficult content, including intricate scientific formulas.\n    *   Chemical structures are intelligently identified and can be represented in the standard **SMILES** format.\n\n*   **Rich, Structured HTML Output**\n    *   The model generates a clean HTML representation of the document, preserving its logical structure.\n    *   Each content block (e.g., paragraph, table, figure, formula) is tagged with its **category**, **bounding box coordinates**, and **OCR text**.\n    *   It automatically identifies and filters out irrelevant elements like headers and footers, focusing only on the core content.\n\n*   **State-of-the-Art Performance**\n    * Logics-Parsing achieves the best performance on our in-house benchmark, which is specifically designed to comprehensively evaluate a model’s parsing capability on complex-layout documents and STEM content.\n\n\n\n\n\n## Benchmark\n\nExisting document-parsing benchmarks often provide limited coverage of complex layouts and STEM content. To address this, we constructed an in-house benchmark comprising 1,078 page-level images across nine major categories and over twenty sub-categories. Our model achieves the best performance on this benchmark.\n\u003cdiv align=\"center\"\u003e\n  \u003cimg src=\"imgs/BenchCls.png\"\u003e\n\u003c/div\u003e\n\u003ctable\u003e\n    \u003ctr\u003e\n        \u003ctd rowspan=\"2\"\u003eModel Type\u003c/td\u003e\n        \u003ctd rowspan=\"2\"\u003eMethods\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eOverall \u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eText Edit \u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eFormula \u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eTable \u003csup\u003eTEDS\u003c/sup\u003e ↑\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eTable \u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd colspan=\"2\"\u003eReadOrder\u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd rowspan=\"1\"\u003eChemistry\u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n        \u003ctd rowspan=\"1\"\u003eHandWriting\u003csup\u003eEdit\u003c/sup\u003e ↓\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eEN\u003c/td\u003e\n        \u003ctd\u003eZH\u003c/td\u003e\n        \u003ctd\u003eALL\u003c/td\u003e\n        \u003ctd\u003eALL\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd rowspan=\"7\"\u003ePipeline Tools\u003c/td\u003e\n        \u003ctd\u003edoc2x\u003c/td\u003e\n        \u003ctd\u003e0.209\u003c/td\u003e\n        \u003ctd\u003e0.188\u003c/td\u003e\n        \u003ctd\u003e0.128\u003c/td\u003e\n        \u003ctd\u003e0.194\u003c/td\u003e\n        \u003ctd\u003e0.377\u003c/td\u003e\n        \u003ctd\u003e0.321\u003c/td\u003e\n        \u003ctd\u003e81.1\u003c/td\u003e\n        \u003ctd\u003e85.3\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.148\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.115\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.146\u003c/td\u003e\n        \u003ctd\u003e0.122\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.307\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eTextin\u003c/td\u003e\n        \u003ctd\u003e0.153\u003c/td\u003e\n        \u003ctd\u003e0.158\u003c/td\u003e\n        \u003ctd\u003e0.132\u003c/td\u003e\n        \u003ctd\u003e0.190\u003c/td\u003e\n        \u003ctd\u003e0.185\u003c/td\u003e\n        \u003ctd\u003e0.223\u003c/td\u003e\n        \u003ctd\u003e76.7\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e86.3\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.176\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.113\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.118\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.104\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.344\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003emathpix\u003csup\u003e*\u003c/sup\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.128\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.146\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.128\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.152\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.06\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.142\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e86.2\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e86.6\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.120\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e0.127\u003c/td\u003e\n        \u003ctd\u003e0.204\u003c/td\u003e\n        \u003ctd\u003e0.164\u003c/td\u003e\n        \u003ctd\u003e0.552\u003c/td\u003e\n        \u003ctd\u003e0.263\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003ePP_StructureV3\u003c/td\u003e\n        \u003ctd\u003e0.220\u003c/td\u003e\n        \u003ctd\u003e0.226\u003c/td\u003e\n        \u003ctd\u003e0.172\u003c/td\u003e\n        \u003ctd\u003e0.29\u003c/td\u003e\n        \u003ctd\u003e0.272\u003c/td\u003e\n        \u003ctd\u003e0.276\u003c/td\u003e\n        \u003ctd\u003e66\u003c/td\u003e\n        \u003ctd\u003e71.5\u003c/td\u003e\n        \u003ctd\u003e0.237\u003c/td\u003e\n        \u003ctd\u003e0.193\u003c/td\u003e\n        \u003ctd\u003e0.201\u003c/td\u003e\n        \u003ctd\u003e0.143\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.382\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eMineru2\u003c/td\u003e\n        \u003ctd\u003e0.212\u003c/td\u003e\n        \u003ctd\u003e0.245\u003c/td\u003e\n        \u003ctd\u003e0.134\u003c/td\u003e\n        \u003ctd\u003e0.195\u003c/td\u003e\n        \u003ctd\u003e0.280\u003c/td\u003e\n        \u003ctd\u003e0.407\u003c/td\u003e\n        \u003ctd\u003e67.5\u003c/td\u003e\n        \u003ctd\u003e71.8\u003c/td\u003e\n        \u003ctd\u003e0.228\u003c/td\u003e\n        \u003ctd\u003e0.203\u003c/td\u003e\n        \u003ctd\u003e0.205\u003c/td\u003e\n        \u003ctd\u003e0.177\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.387\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eMarker\u003c/td\u003e\n        \u003ctd\u003e0.324\u003c/td\u003e\n        \u003ctd\u003e0.409\u003c/td\u003e\n        \u003ctd\u003e0.188\u003c/td\u003e\n        \u003ctd\u003e0.289\u003c/td\u003e\n        \u003ctd\u003e0.285\u003c/td\u003e\n        \u003ctd\u003e0.383\u003c/td\u003e\n        \u003ctd\u003e65.5\u003c/td\u003e\n        \u003ctd\u003e50.4\u003c/td\u003e\n        \u003ctd\u003e0.593\u003c/td\u003e\n        \u003ctd\u003e0.702\u003c/td\u003e\n        \u003ctd\u003e0.23\u003c/td\u003e\n        \u003ctd\u003e0.262\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.50\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003ePix2text\u003c/td\u003e\n        \u003ctd\u003e0.447\u003c/td\u003e\n        \u003ctd\u003e0.547\u003c/td\u003e\n        \u003ctd\u003e0.485\u003c/td\u003e\n        \u003ctd\u003e0.577\u003c/td\u003e\n        \u003ctd\u003e0.312\u003c/td\u003e\n        \u003ctd\u003e0.465\u003c/td\u003e\n        \u003ctd\u003e64.7\u003c/td\u003e\n        \u003ctd\u003e63.0\u003c/td\u003e\n        \u003ctd\u003e0.566\u003c/td\u003e\n        \u003ctd\u003e0.613\u003c/td\u003e\n        \u003ctd\u003e0.424\u003c/td\u003e\n        \u003ctd\u003e0.534\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.95\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd rowspan=\"8\"\u003eExpert VLMs\u003c/td\u003e\n        \u003ctd\u003eDolphin\u003c/td\u003e\n        \u003ctd\u003e0.208\u003c/td\u003e\n        \u003ctd\u003e0.256\u003c/td\u003e\n        \u003ctd\u003e0.149\u003c/td\u003e\n        \u003ctd\u003e0.189\u003c/td\u003e\n        \u003ctd\u003e0.334\u003c/td\u003e\n        \u003ctd\u003e0.346\u003c/td\u003e\n        \u003ctd\u003e72.9\u003c/td\u003e\n        \u003ctd\u003e60.1\u003c/td\u003e\n        \u003ctd\u003e0.192\u003c/td\u003e\n        \u003ctd\u003e0.35\u003c/td\u003e\n        \u003ctd\u003e0.160\u003c/td\u003e\n        \u003ctd\u003e0.139\u003c/td\u003e\n        \u003ctd\u003e0.984\u003c/td\u003e\n        \u003ctd\u003e0.433\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003edots.ocr\u003c/td\u003e\n        \u003ctd\u003e0.186\u003c/td\u003e\n        \u003ctd\u003e0.198\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.115\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.169\u003c/td\u003e\n        \u003ctd\u003e0.291\u003c/td\u003e\n        \u003ctd\u003e0.358\u003c/td\u003e\n        \u003ctd\u003e79.5\u003c/td\u003e\n        \u003ctd\u003e82.5\u003c/td\u003e\n        \u003ctd\u003e0.172\u003c/td\u003e\n        \u003ctd\u003e0.141\u003c/td\u003e\n        \u003ctd\u003e0.165\u003c/td\u003e\n        \u003ctd\u003e0.123\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.255\u003c/ins\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eMonkeyOcr\u003c/td\u003e\n        \u003ctd\u003e0.193\u003c/td\u003e\n        \u003ctd\u003e0.259\u003c/td\u003e\n        \u003ctd\u003e0.127\u003c/td\u003e\n        \u003ctd\u003e0.236\u003c/td\u003e\n        \u003ctd\u003e0.262\u003c/td\u003e\n        \u003ctd\u003e0.325\u003c/td\u003e\n        \u003ctd\u003e78.4\u003c/td\u003e\n        \u003ctd\u003e74.7\u003c/td\u003e\n        \u003ctd\u003e0.186\u003c/td\u003e\n        \u003ctd\u003e0.294\u003c/td\u003e\n        \u003ctd\u003e0.197\u003c/td\u003e\n        \u003ctd\u003e0.180\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.623\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eOCRFlux\u003c/td\u003e\n        \u003ctd\u003e0.252\u003c/td\u003e\n        \u003ctd\u003e0.254\u003c/td\u003e\n        \u003ctd\u003e0.134\u003c/td\u003e\n        \u003ctd\u003e0.195\u003c/td\u003e\n        \u003ctd\u003e0.326\u003c/td\u003e\n        \u003ctd\u003e0.405\u003c/td\u003e\n        \u003ctd\u003e58.3\u003c/td\u003e\n        \u003ctd\u003e70.2\u003c/td\u003e\n        \u003ctd\u003e0.358\u003c/td\u003e\n        \u003ctd\u003e0.260\u003c/td\u003e\n        \u003ctd\u003e0.191\u003c/td\u003e\n        \u003ctd\u003e0.156\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.284\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eGotocr\u003c/td\u003e\n        \u003ctd\u003e0.247\u003c/td\u003e\n        \u003ctd\u003e0.249\u003c/td\u003e\n        \u003ctd\u003e0.181\u003c/td\u003e\n        \u003ctd\u003e0.213\u003c/td\u003e\n        \u003ctd\u003e0.231\u003c/td\u003e\n        \u003ctd\u003e0.318\u003c/td\u003e\n        \u003ctd\u003e59.5\u003c/td\u003e\n        \u003ctd\u003e74.7\u003c/td\u003e\n        \u003ctd\u003e0.38\u003c/td\u003e\n        \u003ctd\u003e0.299\u003c/td\u003e\n        \u003ctd\u003e0.195\u003c/td\u003e\n        \u003ctd\u003e0.164\u003c/td\u003e\n        \u003ctd\u003e0.969\u003c/td\u003e\n        \u003ctd\u003e0.446\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eOlmocr\u003c/td\u003e\n        \u003ctd\u003e0.341\u003c/td\u003e\n        \u003ctd\u003e0.382\u003c/td\u003e\n        \u003ctd\u003e0.125\u003c/td\u003e\n        \u003ctd\u003e0.205\u003c/td\u003e\n        \u003ctd\u003e0.719\u003c/td\u003e\n        \u003ctd\u003e0.766\u003c/td\u003e\n        \u003ctd\u003e57.1\u003c/td\u003e\n        \u003ctd\u003e56.6\u003c/td\u003e\n        \u003ctd\u003e0.327\u003c/td\u003e\n        \u003ctd\u003e0.389\u003c/td\u003e\n        \u003ctd\u003e0.191\u003c/td\u003e\n        \u003ctd\u003e0.169\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.294\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eSmolDocling\u003c/td\u003e\n        \u003ctd\u003e0.657\u003c/td\u003e\n        \u003ctd\u003e0.895\u003c/td\u003e\n        \u003ctd\u003e0.486\u003c/td\u003e\n        \u003ctd\u003e0.932\u003c/td\u003e\n        \u003ctd\u003e0.859\u003c/td\u003e\n        \u003ctd\u003e0.972\u003c/td\u003e\n        \u003ctd\u003e18.5\u003c/td\u003e\n        \u003ctd\u003e1.5\u003c/td\u003e\n        \u003ctd\u003e0.86\u003c/td\u003e\n        \u003ctd\u003e0.98\u003c/td\u003e\n        \u003ctd\u003e0.413\u003c/td\u003e\n        \u003ctd\u003e0.695\u003c/td\u003e\n        \u003ctd\u003e1.0\u003c/td\u003e\n        \u003ctd\u003e0.927\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003e\u003cb\u003eLogics-Parsing\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.124\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.145\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.089\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.139\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.106\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.165\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e76.6\u003c/td\u003e\n        \u003ctd\u003e79.5\u003c/td\u003e\n        \u003ctd\u003e0.165\u003c/td\u003e\n        \u003ctd\u003e0.166\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.136\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.113\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.519\u003c/b\u003e\u003c/td\u003e\n        \u003ctd\u003e\u003cb\u003e0.252\u003c/b\u003e\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd rowspan=\"5\"\u003eGeneral VLMs\u003c/td\u003e\n        \u003ctd\u003eQwen2VL-72B\u003c/td\u003e\n        \u003ctd\u003e0.298\u003c/td\u003e\n        \u003ctd\u003e0.342\u003c/td\u003e\n        \u003ctd\u003e0.142\u003c/td\u003e\n        \u003ctd\u003e0.244\u003c/td\u003e\n        \u003ctd\u003e0.431\u003c/td\u003e\n        \u003ctd\u003e0.363\u003c/td\u003e\n        \u003ctd\u003e64.2\u003c/td\u003e\n        \u003ctd\u003e55.5\u003c/td\u003e\n        \u003ctd\u003e0.425\u003c/td\u003e\n        \u003ctd\u003e0.581\u003c/td\u003e\n        \u003ctd\u003e0.193\u003c/td\u003e\n        \u003ctd\u003e0.182\u003c/td\u003e\n        \u003ctd\u003e0.792\u003c/td\u003e\n        \u003ctd\u003e0.359\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eQwen2.5VL-72B\u003c/td\u003e\n        \u003ctd\u003e0.233\u003c/td\u003e\n        \u003ctd\u003e0.263\u003c/td\u003e\n        \u003ctd\u003e0.162\u003c/td\u003e\n        \u003ctd\u003e0.24\u003c/td\u003e\n        \u003ctd\u003e0.251\u003c/td\u003e\n        \u003ctd\u003e0.257\u003c/td\u003e\n        \u003ctd\u003e69.6\u003c/td\u003e\n        \u003ctd\u003e67\u003c/td\u003e\n        \u003ctd\u003e0.313\u003c/td\u003e\n        \u003ctd\u003e0.353\u003c/td\u003e\n        \u003ctd\u003e0.205\u003c/td\u003e\n        \u003ctd\u003e0.204\u003c/td\u003e\n        \u003ctd\u003e0.597\u003c/td\u003e\n        \u003ctd\u003e0.349\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eDoubao-1.6\u003c/td\u003e\n        \u003ctd\u003e0.188\u003c/td\u003e\n        \u003ctd\u003e0.248\u003c/td\u003e\n        \u003ctd\u003e0.129\u003c/td\u003e\n        \u003ctd\u003e0.219\u003c/td\u003e\n        \u003ctd\u003e0.273\u003c/td\u003e\n        \u003ctd\u003e0.336\u003c/td\u003e\n        \u003ctd\u003e74.9\u003c/td\u003e\n        \u003ctd\u003e69.7\u003c/td\u003e\n        \u003ctd\u003e0.180\u003c/td\u003e\n        \u003ctd\u003e0.288\u003c/td\u003e\n        \u003ctd\u003e0.171\u003c/td\u003e\n        \u003ctd\u003e0.148\u003c/td\u003e\n        \u003ctd\u003e0.601\u003c/td\u003e\n        \u003ctd\u003e0.317\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eGPT-5\u003c/td\u003e\n        \u003ctd\u003e0.242\u003c/td\u003e\n        \u003ctd\u003e0.373\u003c/td\u003e\n        \u003ctd\u003e0.119\u003c/td\u003e\n        \u003ctd\u003e0.36\u003c/td\u003e\n        \u003ctd\u003e0.398\u003c/td\u003e\n        \u003ctd\u003e0.456\u003c/td\u003e\n        \u003ctd\u003e67.9\u003c/td\u003e\n        \u003ctd\u003e55.8\u003c/td\u003e\n        \u003ctd\u003e0.26\u003c/td\u003e\n        \u003ctd\u003e0.397\u003c/td\u003e\n        \u003ctd\u003e0.191\u003c/td\u003e\n        \u003ctd\u003e0.28\u003c/td\u003e\n        \u003ctd\u003e0.88\u003c/td\u003e\n        \u003ctd\u003e0.46\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n        \u003ctd\u003eGemini2.5 pro\u003c/td\u003e\n        \u003ctd\u003e0.185\u003c/td\u003e\n        \u003ctd\u003e0.20\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.115\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.155\u003c/td\u003e\n        \u003ctd\u003e0.288\u003c/td\u003e\n        \u003ctd\u003e0.326\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e82.6\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e80.3\u003c/td\u003e\n        \u003ctd\u003e0.154\u003c/td\u003e\n        \u003ctd\u003e0.182\u003c/td\u003e\n        \u003ctd\u003e0.181\u003c/td\u003e\n        \u003ctd\u003e0.136\u003c/td\u003e\n        \u003ctd\u003e\u003cins\u003e0.535\u003c/ins\u003e\u003c/td\u003e\n        \u003ctd\u003e0.26\u003c/td\u003e\n    \u003c/tr\u003e\n\n\u003c/table\u003e\n\u003c!-- 脚注说明 --\u003e\n\u003ctr\u003e\n  \u003ctd colspan=\"5\"\u003e\n    \u003csup\u003e*\u003c/sup\u003e Tested on the v3/PDF Conversion API (August 2025 deployment).\n\n  \u003c/td\u003e\n\u003c/tr\u003e\n\n\n## Quick Start\n### 1. Installation\n```shell\nconda create -n logis-parsing python=3.10\nconda activate logis-parsing\n\npip install -r requirement.txt\n\n```\n### 2. Download Model Weights\n\n```\n# Download our model from Modelscope.\npip install modelscope\npython download_model.py -t modelscope\n\n# Download our model from huggingface.\npip install huggingface_hub\npython download_model.py -t huggingface\n```\n\n### 3. Inference\n```shell\npython3 inference.py --image_path PATH_TO_INPUT_IMG --output_path PATH_TO_OUTPUT --model_path PATH_TO_MODEL\n```\n\n## Acknowledgments\n\n\nWe would like to acknowledge the following open-source projects that provided inspiration and reference for this work:\n- [Qwen2.5-VL](https://github.com/QwenLM/Qwen2.5-VL)\n- [OmniDocBench](https://github.com/opendatalab/OmniDocBench)\n- [Mathpix](https://mathpix.com/)\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falibaba%2FLogics-Parsing","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Falibaba%2FLogics-Parsing","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falibaba%2FLogics-Parsing/lists"}