LlamaParse vs MinerU
Pick LlamaParse if you want a hosted API with a free tier and a price you can tune per page, mainly for RAG ingestion of mixed PDF and office files. Pick MinerU if you need to self-host, run on CPU, or convert Chinese and English academic papers into Markdown, HTML or LaTeX. Avoid both for bank statements unless you validate the output, since each has a reported failure on them.
LlamaParsellamaindex/llamaparse | MinerUopendatalab/mineru | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 5 credits / page | 2 credits / page |
| Per 1,000 pages | $5.00 | $2.00 |
| Typical speed | 3.0 s / page | 2.0 s / page |
| Best for | RAG ingestion | Academic PDFs |
| Inputs | PDF, images, DOCX | PDF, images |
| Languages | 80+ languages | 80+ languages |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| 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 LlamaParse if
- You want a hosted API with no GPU or server to run, plus 10,000 free credits per month.
- You need to pick a cost and quality tier per job, from Fast at 1 credit per page to Agentic Plus at 45 credits per page.
- You are building RAG ingestion inside a LlamaIndex pipeline and want Markdown or JSON from mixed PDF and office files.
Choose MinerU if
- You must keep documents on your own hardware, since MinerU is self-hosted and free apart from compute.
- You have no GPU: the basic ONNX tier runs on CPU with 2 GB RAM, and the VLM tiers need a GPU with 8 GB+ VRAM when you want more quality.
- You need more output formats than Markdown and JSON, such as HTML, LaTeX, DOCX and EPUB, or you work with Chinese and English academic papers.
Key differences
| LlamaParse | MinerU | |
|---|---|---|
| How you run it | Hosted API | Self-host as CLI, API or WebUI |
| Licence | Proprietary | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights tagged AGPL-3.0 |
| Price | $1.25 per 1,000 credits; about $1.25 to $56 per 1,000 pages depending on tier; 10K free credits per month | Free; you pay for compute |
| Output formats | Markdown, JSON, plain text, images | Markdown, HTML, LaTeX, DOCX, EPUB, JSON |
| Hardware | None on your side | Basic tier: CPU, 2 GB RAM. VLM tiers: 16 GB RAM and a GPU with 8 GB+ VRAM |
| Benchmarks | Not published | olmOCR-Bench overall 75.2 (MinerU 2.5.4); ParseBench mean 45.9 (MinerU2.5) |
| Known failure | Invented transactions on bank statements in Accurate mode (issue #450); hallucination on low-quality scans with no confidence score (issue #483) | Repeated content producing 300 KB figure blocks (issue #5728); merged header rows dropped in tables such as bank statements (issue #5554) |
Sources
- LlamaIndex pricing
- GitHub issue #450
- GitHub issue #483
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
What each one is
MinerU
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.
PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?
Use one, fall back to the other
{"model": "llamaindex/llamaparse","fallbacks": ["opendatalab/mineru"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
LlamaParse vs DoclingLlamaParse vs Unlimited OCRMathpix Convert vs MinerUClaude vs MinerULlamaParse vs Reducto ParseMinerU vs DoclingLlamaParse vs Tesseract.jsMinerU vs Phi-4-multimodal
LlamaParse alternativesMinerU alternativesBest OCR models in 2026