- Go 100%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| .github/workflows | ||
| .memory-bank | ||
| examples/parse-file | ||
| pkg | ||
| references | ||
| .gitignore | ||
| .markdownlint-cli2.jsonc | ||
| base.go | ||
| config.go | ||
| constants.go | ||
| content_types.go | ||
| COPYING.md | ||
| create.go | ||
| files.go | ||
| go.mod | ||
| go.sum | ||
| jobs.go | ||
| LICENSE | ||
| mineru.go | ||
| nullable.go | ||
| nullable_test.go | ||
| README.md | ||
| response.go | ||
| SECURITY.md | ||
| usage.go | ||
| version.go | ||
GMS MinerU Client Library
Introduction
MinerU is a document extraction, deep learning document analysis and conversion framework, which can also be run as a service.
This package is a framework for interfacing Go based applications with the API presented by the OpenAPI services made available through the MinerU Router and standalone API services. The service implements an OpenAPI structured API however generated API libraries in Go tend to be cumbersome and this package aims to simplify interactions by implementing a concise Go based interface to the presented OpenAPI structured endpoints.
MinerU Reference Documentation
The references folder contains useful documentation cherry picked from the MinerU project and if needed tweaked to ensure it is readable in English as a good portion of the MinerU documentation is either in Chinese or has be roughly translated from Chinese. As such the documentation in this folder may differ slightly from the original in tone or grammar as it has undergone a English Technical Author proof read to remove any potentially ambiguous elements.
API Versions
There are two API versions handled by this package, the initial MinerU 3.x.x based API is represented as the mineru-legacy API within the package. This element of the package is likely only see minor updates to fix bugs and package breaking issues. The core of the package has moved to supper the new MinerU 4.x.x based V1 API, which is currently under development. The new API is a significantly different approach and is not compatible with the older now legacy API.
Features
- Formulas - LaTeX - Tables - HTML, accurate complex layout reconstruction
- Supports scanned docs, handwriting, multi-column layouts, cross-page table merging
- Output follows human reading order with automatic header/footer removal
- VLM + OCR dual engine, 109-language OCR recognition
- Leverages the Self-Hosted MinerU 4.x V1 API
Copyright and License
Code and Documentation Copyright © 1994-2026, Stephen Kapp and Reaper Technologies Limited. All Rights Reserved.