MinerU vs Surya
Pick MinerU for mixed corpora of PDFs and Office files, especially Chinese and English academic papers, or for CPU-only batch conversion. Pick Surya for local OCR with layout detection, reading order and bounding boxes across 90+ languages. The main split is input and output range: MinerU converts many formats to Markdown, DOCX or EPUB, while Surya returns structured blocks with coordinates from a smaller 689M model.
MinerUopendatalab/mineru | Suryadatalab/surya | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 2.0 s / page | 1.5 s / page |
| Best for | Academic PDFs | Layout and line-level OCR |
| Inputs | PDF, images | PDF, images |
| Languages | 80+ languages | 90+ languages |
| 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 | ✓ 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 MinerU if
- It accepts PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool, which suits mixed RAG corpora.
- It renders to Markdown, HTML, LaTeX, DOCX, EPUB and JSON, so you can convert straight into the format you need.
- Its basic ONNX tier runs on CPU with about 2 GB RAM, so you can start batch conversion without a GPU.
Choose Surya if
- It returns layout labels, reading order, OCR, HTML tables and KaTeX math with per-block bounding boxes from one model.
- It scores 83.7% on olmOCR-bench with 689M parameters (reported by the authors), against 75.2 for MinerU 2.5.4 on the same benchmark.
- It runs on a laptop via llama.cpp on CPU or Apple Silicon, with peak memory of about 4.3 to 5.9 GB on an M4 Max.
Key differences
| MinerU | Surya | |
|---|---|---|
| Licence | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights tagged AGPL-3.0 | Code Apache-2.0; weights under a modified AI Pubs Open Rail-M, free for research, personal use and companies under $5M funding or revenue |
| How you run it | Local CLI, API or WebUI, from a CPU-only tier to VLM tiers | Lightweight client against a vLLM or llama.cpp inference server you run yourself |
| Hardware | Basic tier: CPU, 2 GB RAM. VLM tiers: 16 GB RAM and a GPU with 8 GB+ VRAM | NVIDIA GPU via vLLM, or CPU/Apple Silicon via llama.cpp |
| Input formats | PDF, images, DOCX, PPTX, XLSX, EPUB, OFD, HTML | Not published (not meant for photos or natural scenes) |
| Output formats | Markdown, HTML, LaTeX, DOCX, EPUB, JSON | JSON, HTML, LaTeX, bounding boxes, plain text |
| olmOCR-Bench | 75.2 overall (MinerU 2.5.4, independent) | 83.7% (Surya 2.5, reported by the authors) |
| Known failure | Drops merged header rows in tables such as bank statement account lines; VLM pipeline can repeat content into 300 KB figure blocks | Surya 2.5 table output lacks row and column geometry; multilingual tables score 58.1; the README sends handwriting and complex tables to the hosted Datalab API |
Sources
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
- Issue #5728 repetition
- Surya GitHub README
- Datalab pricing and license FAQ
What each one is
MinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU detailsSurya
Open-source OCR toolkit with line detection, layout analysis, reading order and tables.
Surya 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": ["datalab/surya"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
MinerU alternativesSurya alternativesBest OCR models in 2026