MinerU
Document parser that converts PDFs, scans and Office files into clean Markdown or JSON for LLMs and RAG.
Quick answer
MinerU is a document parser that converts PDFs, scans and Office files into clean Markdown or JSON for LLMs and RAG, listed under Document processing. A typical small-business use: turn scanned contracts, invoices or manuals into searchable text before loading them into a RAG app. Official hardware: minimum 2 GB RAM. That fits a Hetzner Cloud CPX12 server at about €11.99/month (server only, excl. VAT). License: MinerU Open Source License (Apache-2.0 with additional terms), which allows internal business use.
Last updated · facts verified against official docs on · sources
Activity data comes from GitHub and was last refreshed Oct 22, 2026.
What is MinerU?
MinerU parses complex documents (multi-column layouts, tables, formulas, scanned pages) into Markdown or JSON that a knowledge base or LLM can use. It ships a web UI, a command-line tool and a V1 parsing API, with parsing tiers from fast text-layer extraction to vision-language-model parsing. The base package runs small ONNX models on CPU; NVIDIA GPUs give higher throughput.
Where it fits in a 10–200 person business
- Turn scanned contracts, invoices or manuals into searchable text before loading them into a RAG app
- Batch-convert a folder of PDFs and Word files to Markdown for an internal knowledge base
- Extract tables from reports without sending the files to a cloud OCR service
These use cases are our editorial suggestions, not claims made by the project.
Can a business use MinerU? (license)
| License | MinerU Open Source License (Apache-2.0 with additional terms) |
|---|---|
| Type | Custom / conditional: Apache- or BSD-based with extra conditions (branding, multi-tenant SaaS, derivatives). |
| Internal business use | Allowed under the license text |
| What to watch | Commercial use is allowed without a separate license, unless you and your affiliates exceed 100 million monthly active users or USD 20 million monthly revenue. If you offer an online service to third parties based on MinerU, you must state prominently that it uses MinerU. |
This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.
Data & compliance: does MinerU phone home?
| Telemetry (usage data sent to the vendor) | On by default May collect anonymous, locally aggregated usage and diagnostic telemetry (no document content, file names or prompts); uploads stop only when you disable it explicitly. How to turn it off: Run mineru telemetry disable. |
|---|---|
| Other calls home by default (updates, marketplaces, default cloud services) | Not documented |
| Single sign-on (OIDC / SAML / LDAP) | Not documented |
| Audit log | Not documented |
| Role-based access control (RBAC) | Not documented |
| Data residency / GDPR notes from the project | Not documented |
Checked against the project's own docs and repository on . "Not documented" means we could not find the answer in official sources; it does not mean the feature is missing or that no data is sent. Third-party services you connect (model APIs, vector databases, search engines) have their own data policies. Compare all apps.
What hardware does MinerU need?
| Minimum | 2 GB RAM |
|---|---|
| Recommended | Not published |
| GPU | Optional. The basic tier runs ONNX models on CPU; the standard/advanced tiers work on CPU with 8 GB RAM (Vulkan recommended). PyTorch and vLLM/lmdeploy engines need a GPU with 8 GB+ VRAM and 16 GB RAM. The official Docker image targets NVIDIA GPUs only. |
The 2 GB figure is the official minimum RAM for the basic (ONNX, CPU) tier; higher-quality tiers need 8–16 GB.
Source: github.com/opendatalab/MinerU#model-engines-with-extras
How much does it cost to host MinerU per month?
| Minimum spec | 2 GB RAM → CPX12 (1 vCPU, 2 GB, 40 GB disk) €11.99/mo |
|---|
How we got this: we took the cheapest Hetzner Cloud Regular Performance (shared AMD vCPU) plan with at least the RAM and vCPU the project lists, using prices read from www.hetzner.com/cloud/regular-performance on 2026-10-22. "From" monthly price cap shown by Hetzner, excluding VAT. Prices vary by location. Covers the server only; it excludes LLM API usage, backups, extra storage, GPUs and your time. Disk is not matched; add a volume if you need more than the plan includes.
How to deploy MinerU
Method: Python package (uv/pip) with web UI; Docker Compose for NVIDIA GPUs
uv venv --python 3.12 .venv
source .venv/bin/activate
uv pip install -U "mineru>=4.0,<5"
mineru-kit webui --server-name 127.0.0.1 --server-port 7860
# API instead: mineru-kit api-server --host 127.0.0.1 --port 8000 --tier standardFrom the official instructions: opendatalab.github.io/MinerU/quick_start/. Always check the source for the current version before you run it.
Known gotchas when self-hosting MinerU
- Docker deployment in 4.0 is for Linux with NVIDIA GPUs only; non-NVIDIA Docker setups are pending an update and their legacy Dockerfiles stay on MinerU below 4.0. source
- The release Dockerfile installs the version range mineru[torch]>=4.0,<5 from the package index, so rebuilding the same tag later can give a different 4.x version. Tag images with the actual version and check it after building. source
- The flash tier with --ocr-mode txt only reads the PDF text layer and does not OCR scanned pages; use --ocr-mode ocr for scans. source
- Running a public parsing service for others requires a visible "uses MinerU" notice under the license. source
Alternatives in Document processing
- Paperless-AI AI auto-tagging and document chat for Paperless-ngx (currently unmaintained). Compare MinerU vs Paperless-AI
MinerU FAQ
How much RAM does MinerU need?
The official minimum is 2 GB RAM. The 2 GB figure is the official minimum RAM for the basic (ONNX, CPU) tier; higher-quality tiers need 8–16 GB.
Does MinerU need a GPU?
Optional. The basic tier runs ONNX models on CPU; the standard/advanced tiers work on CPU with 8 GB RAM (Vulkan recommended). PyTorch and vLLM/lmdeploy engines need a GPU with 8 GB+ VRAM and 16 GB RAM. The official Docker image targets NVIDIA GPUs only.
How much does it cost to self-host MinerU?
The software is free to download. Based on the official minimum 2 GB RAM: That fits a Hetzner Cloud CPX12 server at about €11.99/month (server only, excl. VAT). That covers the server only, not LLM API usage, backups, extra storage or your time.
Is MinerU free for commercial use?
MinerU is licensed under MinerU Open Source License (Apache-2.0 with additional terms) (Custom / conditional). Internal business use is allowed under the license text. Commercial use is allowed without a separate license, unless you and your affiliates exceed 100 million monthly active users or USD 20 million monthly revenue. If you offer an online service to third parties based on MinerU, you must state prominently that it uses MinerU. This is a summary, not legal advice.
What are common problems when self-hosting MinerU?
Docker deployment in 4.0 is for Linux with NVIDIA GPUs only; non-NVIDIA Docker setups are pending an update and their legacy Dockerfiles stay on MinerU below 4.0. The release Dockerfile installs the version range mineru[torch]>=4.0,<5 from the package index, so rebuilding the same tag later can give a different 4.x version. Tag images with the actual version and check it after building. The flash tier with --ocr-mode txt only reads the PDF text layer and does not OCR scanned pages; use --ocr-mode ocr for scans. Running a public parsing service for others requires a visible "uses MinerU" notice under the license.
Is MinerU still maintained?
Yes, it is actively developed. Last commit: Oct 11, 2026. Latest release: mineru-4.0.12-released (Oct 11, 2026). Checked Oct 22, 2026.
How do I install MinerU?
The official method is Python package (uv/pip) with web UI; Docker Compose for NVIDIA GPUs. The exact commands are in the "How to deploy" section above and in the official instructions: https://opendatalab.github.io/MinerU/quick_start/
What are self-hosted alternatives to MinerU?
Other apps in Document processing: Paperless-AI.
Sources
- GitHub repository: github.com/opendatalab/MinerU
- Official documentation, Deploy instructions, Gotcha 3: opendatalab.github.io/MinerU/quick_start/
- License text, Gotcha 4: github.com/opendatalab/MinerU/blob/master/LICENSE.md
- Hardware requirements: github.com/opendatalab/MinerU#model-engines-with-extras
- Gotcha 1, Gotcha 2: opendatalab.github.io/MinerU/quick_start/docker_deployment/
- Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
- GitHub stars, commits and releases: GitHub REST API, fetched
Facts on this page were checked against the project's own repository and docs on 2026-10-22. Spotted something out of date? Tell us.