MinerU vs DeepSeek-VL2
Pick MinerU to convert mixed PDF, image and Office files into Markdown, HTML, LaTeX, DOCX or JSON, especially Chinese and English academic papers or CPU-only batch jobs. Pick DeepSeek-VL2 to answer questions about a single document page or chart, where it scores DocVQA 93.3 and ChartQA 86.0. The main split is full-document conversion versus per-page question answering.
MinerUopendatalab/mineru | DeepSeek-VL2deepseek/deepseek-vl2 | |
|---|---|---|
| Type | Open source | Vision model |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 2.0 s / page | 3.0 s / page |
| Best for | Academic PDFs | Charts and tables |
| Inputs | PDF, images | images |
| Languages | 80+ languages | Mainly English and Chinese |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | ✓ Yes | No |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
Which to choose
Choose MinerU if
- It accepts PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool, so a mixed corpus needs no extra converters.
- It exports to Markdown, HTML, LaTeX, DOCX, EPUB and JSON, which suits RAG pipelines and publishing.
- Its basic tier runs on CPU with about 2 GB RAM, so you can start without a GPU.
Choose DeepSeek-VL2 if
- It scores 93.3 on DocVQA test and 86.0 on ChartQA test, which fits document and chart question answering.
- The Tiny model runs under 40GB of GPU memory, and it activates only 1.0B to 4.5B parameters per token.
- The weights allow commercial use under the DeepSeek Model License, and the code is MIT.
Key differences
| MinerU | DeepSeek-VL2 | |
|---|---|---|
| Main job | Parses PDFs, images and Office files into Markdown or JSON | Vision language model for document and chart question answering and single page understanding |
| Licence | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights tagged AGPL-3.0 | MIT for code; DeepSeek Model License for weights, commercial use allowed |
| Hardware | Basic tier: CPU with 2 GB RAM. VLM tiers: 16 GB RAM and a GPU with 8 GB+ VRAM | Small may need 80GB GPU memory (40GB with incremental prefilling); Tiny under 40GB |
| Model size | MinerU2.5 VLM: 1.2B | Tiny 3.37B total (1.0B active), Small 16.1B, full 27.5B total (4.5B active) |
| Output formats | Markdown, HTML, LaTeX, DOCX, EPUB, JSON | Plain text, Markdown |
| Published benchmarks | olmOCR-Bench overall 75.2 (MinerU 2.5.4); ParseBench mean 45.9 (MinerU2.5) | DocVQA test 93.3; OCRBench 834 (Small); ChartQA test 86.0, all from the authors |
| Known failure | VLM tiers drop merged header rows in tables such as bank statement account lines; the VLM pipeline can generate repeated content, including 300 KB figure blocks | Struggles with blurry images; handles only a few images per session, and dynamic tiling turns off with more than two images |
Sources
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
- DeepSeek-VL2 GitHub README
- Paper arXiv 2412.10302
What each one is
MinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU detailsDeepSeek-VL2
DeepSeek's mixture-of-experts vision models (December 2024) with OCR and table and chart understanding.
DeepSeek-VL2 detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "opendatalab/mineru","fallbacks": ["deepseek/deepseek-vl2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
MinerU alternativesDeepSeek-VL2 alternativesBest OCR models in 2026