GOT-OCR 2.0 vs Unstructured
Pick GOT-OCR 2.0 for self-hosted reading of formulas, sheet music, chemical structures and boxed regions in English or Chinese documents. Pick Unstructured to ingest mixed file types such as PDFs, Word, PowerPoint, HTML and email into typed elements for RAG. The main split is scope: GOT-OCR 2.0 is a single OCR model, while Unstructured is a document pipeline with a free library and a hosted API.
GOT-OCR 2.0stepfun/got-ocr-2 | Unstructuredunstructured/unstructured | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 1 credit / page | 20 credits / page |
| Per 1,000 pages | $1.00 | $20.00 |
| Typical speed | 1.5 s / page | 3.0 s / page |
| Best for | Formatted text, formulas and charts | Mixed file types into RAG |
| Inputs | images | PDF, images, DOCX |
| Languages | Chinese, English | Multilingual |
| Max pages per request | 100 | 1,000 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | No | ✓ 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 GOT-OCR 2.0 if
- It reads formulas, sheet music, chemical structures, charts and geometry, and outputs LaTeX, TikZ, SMILES and kern.
- It supports region OCR guided by box coordinates or color, so you can read one part of a page.
- It is a 580M-parameter model you run yourself, with no per-page fee beyond your compute.
Choose Unstructured if
- It handles PDFs, HTML, Word, PowerPoint, emails and images in one library, with many connectors.
- The hosted platform gives you the first 10,000 pages free, then $0.015 per page, so you can start without running GPUs.
- Its Auto, Fast, High-Res and VLM strategies let you trade speed for quality, and Fast suits born-digital PDFs.
Key differences
| GOT-OCR 2.0 | Unstructured | |
|---|---|---|
| What it is | 580M-parameter end-to-end OCR model for text, formulas, tables, charts and sheet music | Document ETL library and hosted platform that partitions files into typed elements |
| How you run it | Self-hosted only. CUDA 11.8 and PyTorch 2.0.1 per README; CPU users are pointed to a llama.cpp port | Self-host the open-source library or call the hosted API |
| Licence | Code Apache-2.0; training data CC BY-NC 4.0 | Library Apache-2.0; hosted platform proprietary |
| Price | Free, you pay for compute | Library free. Hosted: first 10,000 pages free, then $0.015 per page; Business is custom |
| Output formats | Plain text, Markdown, LaTeX, TikZ, SMILES, kern | JSON, HTML, plain text, bounding boxes |
| Benchmark | Not published; the README lists no benchmark scores | Table cell content: 0.426 open-source vs 0.866 hosted, from the authors |
| Known failure | Output can loop characters midway (issue #226), return wrong text on images with no text (issue #235), and format mode is much worse than plain OCR mode (issue #317) | hi_res tables can cut off at edges and drop first letters (issue #3262), and table OCR ignores language settings (issue #1694) |
Sources
What each one is
GOT-OCR 2.0
Unified OCR model that reads plain text, tables, formulas, sheet music and charts.
GOT-OCR 2.0 detailsUnstructured
Document ETL for RAG. Partitions many file types into typed elements; open-source library too.
Unstructured 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": "stepfun/got-ocr-2","fallbacks": ["unstructured/unstructured"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
GOT-OCR 2.0 alternativesUnstructured alternativesBest OCR models in 2026