MinerU vs Chandra
Pick MinerU if you need one tool for PDFs plus Office files, or a CPU-only batch job with no GPU. Pick Chandra if you need stronger multilingual OCR on scans, or a hosted API with per-page pricing. The main split is hardware and licence: MinerU starts on 2 GB RAM with a custom licence, while Chandra needs a GPU and limits free commercial self-hosting to companies under $2M funding or revenue.
MinerUopendatalab/mineru | Chandradatalab/chandra | |
|---|---|---|
| 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 | 2.0 s / page |
| Best for | Academic PDFs | Complex layouts and handwriting |
| Inputs | PDF, images | PDF, images |
| Languages | 80+ languages | 40+ languages |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | ✓ Yes |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | No | ✓ Yes |
| 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 corpora for RAG.
- Its basic ONNX tier runs on CPU with about 2 GB RAM, so you can start without a GPU.
- It can render to Markdown, HTML, LaTeX, DOCX and EPUB, and it suits Chinese and English academic papers.
Choose Chandra if
- It scores 85.8 on olmOCR-bench (Chandra 2, authors' table) against 75.2 for MinerU 2.5.4 on the same benchmark.
- It averages 77.8% across 43 languages, which fits multilingual scanned documents.
- You can self-host or call the Datalab API at $1, $3 or $7.50 per 1,000 pages, with a $20/month free tier.
Key differences
| MinerU | Chandra | |
|---|---|---|
| How you run it | Self-host only: CLI, API or WebUI | Self-host (vLLM or Hugging Face) or hosted Datalab API |
| 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 modified OpenRAIL-M, free under $2M funding/revenue, commercial licence above that |
| Hardware | Basic tier: CPU, 2 GB RAM. VLM tiers: 16 GB RAM and a GPU with 8 GB+ VRAM | Single NVIDIA GPU via vLLM; presets for A100, L40S and RTX 4090/5090 |
| Price | Free, you pay for compute | Self-hosted free within licence terms; API $1, $3 or $7.50 per 1,000 pages |
| Input and output | PDF, images, Office files, EPUB, OFD, HTML in; Markdown, HTML, LaTeX, DOCX, EPUB, JSON out | Pages in; Markdown, HTML, JSON and bounding boxes out |
| olmOCR-Bench overall | 75.2 (MinerU 2.5.4, independent) | 85.8 (Chandra 2, reported by authors) |
| Known failure | VLM tiers drop merged header rows in tables such as bank statement account lines; repeated content can produce 300 KB figure blocks | Token-loop hallucinations on some Russian/Kazakh scans; weak on Urdu (43.2%) and Sanskrit (51.1%); Old Scans only 51.1 |
Sources
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
- Chandra GitHub README
- Datalab pricing
- GitHub issue #51 token loops
What each one is
MinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU detailsChandra
Datalab's OCR model that keeps layout and handles forms, tables and handwriting.
Chandra 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/chandra"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
MinerU alternativesChandra alternativesBest OCR models in 2026