PaddleOCR-VL vs MinerU
Pick PaddleOCR-VL for multilingual scans, formula-heavy papers and charts on a modest GPU, under a plain Apache-2.0 licence. Pick MinerU for mixed Office and PDF corpora, for output formats like DOCX, LaTeX and EPUB, or when you need a CPU-only tier. The main split is scope: PaddleOCR-VL is one small model for page parsing, MinerU is a toolkit with several quality tiers and a more complicated licence.
PaddleOCR-VLpaddle/paddleocr-vl | MinerUopendatalab/mineru | |
|---|---|---|
| Type | Open source | Open source |
| Price | 6 credits / page | 2 credits / page |
| Per 1,000 pages | $6.00 | $2.00 |
| Typical speed | 1.3 s / page | 2.0 s / page |
| Best for | CJK documents | Academic PDFs |
| Inputs | PDF, images | PDF, images |
| Languages | 109 languages | 80+ languages |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | No | No |
| Multilingual | ✓ Yes | ✓ Yes |
| 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 PaddleOCR-VL if
- It is Apache-2.0 with no extra conditions, so commercial use is straightforward.
- It supports 109 languages in one model, including Cyrillic, Arabic, Devanagari and Thai scripts.
- It has higher published scores: 80.0 on olmOCR-Bench against 75.2 for MinerU 2.5.4, and 96.33 on OmniDocBench v1.6 (reported by the authors).
Choose MinerU if
- It accepts PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool.
- It renders to Markdown, HTML, LaTeX, DOCX, EPUB and JSON.
- Its basic ONNX tier runs on CPU with about 2 GB RAM, so you can start without a GPU.
Key differences
| PaddleOCR-VL | MinerU | |
|---|---|---|
| Licence | Apache-2.0 | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights are tagged AGPL-3.0 |
| Model size | 0.9B parameters | MinerU2.5 VLM: 1.2B parameters |
| Hardware | NVIDIA GPU recommended; CPU possible but slow; no minimum VRAM stated | Basic tier runs on CPU with 2 GB RAM; VLM tiers need 16 GB RAM and a GPU with 8 GB+ VRAM |
| Input and output formats | Outputs Markdown and JSON | Reads PDF, images, Office files, EPUB, OFD and HTML; writes Markdown, HTML, LaTeX, DOCX, EPUB and JSON |
| olmOCR-Bench overall | 80.0 | 75.2 (MinerU 2.5.4) |
| Known failure | Version 1.6 hangs on blurry images, decodes about 11x slower because use_cache ships as false, and regresses Russian and Kazakh OCR on born-digital PDFs | VLM tiers drop merged header rows in tables such as bank statement account lines, and the VLM pipeline can generate repeated content such as 300 KB figure blocks |
| Price | Free, self-hosted (you pay for compute) | Free, self-hosted (you pay for compute) |
Sources
- PaddleOCR-VL-1.6 model card
- olmOCR-Bench table
- MinerU GitHub README
- Issue #18453 use_cache slowdown
- Issue #5554 table header rows
What each one is
PaddleOCR-VL
Open-source document model from Baidu. Excellent on Chinese, Japanese and Korean.
PaddleOCR-VL detailsMinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU 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": "paddle/paddleocr-vl","fallbacks": ["opendatalab/mineru"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
PaddleOCR-VL alternativesMinerU alternativesBest OCR models in 2026