What's Changed

• feat: refactor chart handling in DOCX and PPTX converters to improve extraction and rendering by @myhloli in #4838

Full Changelog: mineru-3.1.2-released...mineru-3.1.3-released

What's Changed

fix: prevent abnormal server termination caused by excessively long PDF rendering time in router mode.

Full Changelog: mineru-3.1.1-released...mineru-3.1.2-released

What's Changed

• fix: Mitigate potential inference hangs on Ascend NPU platforms. by @myhloli in #4821

Full Changelog: mineru-3.1.0-released...mineru-3.1.1-released

What's Changed

• 2026/04/18 3.1.0 Released

  This release focuses on licensing openness, parsing accuracy, and full-format native support. The main updates include:

  • License upgrade
    • MinerU has officially moved from AGPLv3 to the MinerU Open Source License, a custom license based on Apache 2.0.
    • This change significantly reduces adoption friction for both community users and commercial deployments, making MinerU easier to integrate into real-world workflows.
  • VLM main model upgrade
    • The primary VLM model has been upgraded to MinerU2.5-Pro-2604-1.2B, bringing overall parsing accuracy to a state-of-the-art level.
    • The new model now supports image and chart parsing, truncated paragraph merging, cross-page table merging, and image recognition inside tables, further strengthening performance on complex document layouts.
  • Full-format native parsing support
    • Native parsing support has now been extended to PPTX and XLSX.
    • MinerU now fully supports parsing across images, PDF, DOCX, PPTX, and XLSX, providing a more complete multi-format document understanding workflow.

• 2026/04/18 3.1.0 发布

  本次版本更新聚焦于许可协议开放性、解析精度提升与全格式原生支持。主要更新内容包括：

  • 许可协议升级
    • MinerU 已正式从 AGPLv3 切换至基于 Apache 2.0 的 MinerU 开源许可证。
    • 新的许可方式在兼顾开源协作与商业落地的同时，进一步降低了社区使用和商业化接入门槛，让 MinerU 更容易融入真实业务流程。
  • VLM 主模型升级
    • VLM 主模型正式切换为 MinerU2.5-Pro-2604-1.2B，整体解析精度提升至业内领先水平。
    • 新模型现已支持子图切分合并、图像与图表解析、截断段落合并、跨页面表格合并以及表格内图像识别，复杂版面场景下的解析能力进一步增强。
  • 全格式原生解析支持
    • 新增 PPTX 与 XLSX 原生解析能力。
    • 至此，MinerU 已完整支持图片、PDF、DOCX、PPTX、XLSX 全格式解析，为多类型文档统一处理提供了更完整的能力闭环。

Full Changelog: mineru-3.0.9-released...mineru-3.1.0-released

What's Changed

• fix #4742: add function to identify disallowed control Unicode characters by @myhloli in #4743
• fix #4744 by @myhloli in #4745:
  • enhance table merging logic with improved row metrics and state management
  • add aspect ratio checks and character count limits for PDF processing

Full Changelog: mineru-3.0.8-released...mineru-3.0.9-released

What's Changed

• fix: #4728 #4730 implement process management and shutdown mechanisms for MinerU by @myhloli in #4731

Full Changelog: mineru-3.0.7-released...mineru-3.0.8-released

What's Changed

• fix: strip newline characters from paragraph text in office_middle_json_mkcontent by @myhloli in #4717

Full Changelog: mineru-3.0.6-released...mineru-3.0.7-released

What's Changed

#4708:

• feat: add underscore thematic break escaping to Markdown processing
• fix: correct paragraph text extraction by removing unnecessary stripping
• feat: enhance paragraph text extraction to include inline content controls

Full Changelog: mineru-3.0.5-released...mineru-3.0.6-released

What's Changed

• fix: improve shutdown handling for FastAPI child process on Windows at 3.0.4
• fix: add custom JSON schema for file upload in Swagger UI to support fastapi>=0.130.0
• fix: update the sys_platform identifier for Windows in pyproject.toml to resolve the issue where installing [all] on Windows does not automatically install lmdeploy
• feat: add albumentations dependency to pyproject.toml #4701

Full Changelog: mineru-3.0.4-released...mineru-3.0.5-released

What's Changed

• feat: add --enable-vlm-preload option to CLI for VLM model preloading during startup by @myhloli in #4693

Full Changelog: mineru-3.0.3-released...mineru-3.0.4-released