/jump to repo

Repos Recent Bookmarks Watched Notes Tags Discover Compare Stats GitHub

Repository brief

opendatalab/MinerU

Read the upstream summary on the left, browse the cached forks below it, and load each fork comparison into the right-hand panel.

Cached analysis

cached 2026-03-30T11:40:46.087Z

3mo ago

opendatalab/MinerU

MinerU is a Python open source project for converting complex documents, especially PDFs, into LLM-ready Markdown/JSON for agentic workflows. It is active, popular, and forkable: 57,566 stars, 4,766 forks, and recent commits on 2026-03-30. The repo includes CLI tools, API/server entry points, Gradio, Docker, docs, demos, tests, and packaging for PyPI.

Loading tags...

Stars57,566

Forks4,766

Default branchmaster

Last pushed2026-03-30T09:32:12Z

Recommended shortcuts

Jump straight into Discofork's strongest cached fork picks, or open a compare view in one click.

Forks

Choose a fork to inspect

10 of 10 fork briefs

Maintenance:

Magnitude:

Sort:

Selected

Choose this fork only if its bundled web/deployment customizations are the point of adoption and you are prepared to maintain a substantial fork gap. If you want current MinerU capabilities, upstream is the safer choice.

Choose this fork if you want a more turnkey API/Docker/demo packaging of MinerU and can tolerate being far behind upstream. Choose upstream if you care about current fixes, active development, and the full maintained feature set.

Choose this fork if you want a browser-centric MinerU deployment with task persistence and richer result review. Choose upstream if you want the latest project fixes, broader maintenance, and less divergence risk.

Choose the fork only if you explicitly want this older baseline and can live without upstream progress. For most adopters, upstream MinerU is the better choice because it is active and substantially ahead.

Prefer this fork only if you specifically want an older pinned MinerU snapshot. For most adopters, upstream is the better choice because this fork shows no added capabilities and is materially behind recent fixes and deployment work.

Prefer this fork if your priority is deployment convenience, Chinese/local environment support, or a simplified service-oriented setup. Prefer upstream if you want the latest extraction quality, active maintenance, and fewer missing features.

Prefer upstream unless you specifically need this older snapshot. This fork adds no visible capabilities and is far behind current MinerU, so it is mainly useful as a starting point for self-maintained customization.

Choose this fork only if you specifically want its web API/Docker packaging workflow. For most adopters, upstream MinerU is the better base because it is much newer, actively maintained, and far more complete.

Choose this fork only if its language-control and human-in-the-loop extraction workflow are the point. If you want a maintained, upstream-aligned document conversion stack, the fork is too stale and too far diverged to be the safer default.

Prefer upstream unless you specifically need a frozen, older baseline. This fork shows no local feature work and is materially behind, so it is better suited as a snapshot than as the foundation for active adoption.

Fork comparison

papayalove/Magic-PDF

38/100

stale

significant_divergence

Choose this fork only if its bundled web/deployment customizations are the point of adoption and you are prepared to maintain a substantial fork gap. If you want current MinerU capabilities, upstream is the safer choice.

Likely purpose

To maintain a customized MinerU-based document extraction distribution with added web frontend/deployment pieces and local behavior tweaks, rather than to track upstream closely.

Best for

Teams that want to own a downstream MinerU distribution and are willing to maintain a large fork.; Users who need a web-facing PDF extraction workflow bundled with deployment assets.; Integrators who care more about a stable customized codebase than staying current with upstream.; Operators who want regional Docker/deployment customization out of the box.

Additional features

Browser/web UI work under `projects/web`, including a bundled PDF.js viewer and frontend lockfiles.
Deployment workflow additions such as a `mineru-router` service and updated Docker deployment instructions.
China/global Docker environment variants, suggesting regional deployment support.
Hybrid-client backend dependency checks and clearer runtime requirement messaging.
Shutdown handling improvements for local API/STDIN workflows.
Mac-specific concurrency tuning for request handling.

Missing features

It is far behind upstream, so it likely misses many recent MinerU fixes, refinements, and compatibility updates.
It likely lags upstream's current CLI/API/server behavior and newer deployment/router changes.
It likely does not reflect the newest upstream language-guessing, shutdown, and serving improvements.
Upstream is currently active; this fork does not appear to be keeping pace with that release cadence.
It trails upstream by 2744 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

May be easier to adapt if you want a fork that already includes web UI and deployment customization.
Shows attention to operational details like shutdown handling and environment-specific concurrency.
Could be useful for teams who need a bundled downstream distribution rather than a fast-moving upstream dependency.

Risks

Large divergence from upstream makes rebasing and security/update tracking expensive.
High chance of version skew in core extraction behavior, dependencies, and bundled assets.
Because it is stale relative to upstream, adopters may inherit old bugs or miss newer fixes and model/runtime compatibility work.
Fork-specific deployment and web pieces may complicate maintenance if you only need the upstream CLI/API.

opendatalab/MinerU · Discofork