docTR vs OpenOCR
Pick docTR for word-level OCR with coordinates, such as invoices and receipts that feed your own extraction code, or if your team trains its own models in Python. Pick OpenOCR for Chinese and English academic PDFs with formulas, where you want document parsing from a 0.1B model. The main split is output: docTR returns words and lines only, while OpenOCR covers formulas, tables and document parsing.
docTRmindee/doctr | OpenOCRfudan/openocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 1 credit / page |
| Per 1,000 pages | $1.00 | $1.00 |
| Typical speed | 0.9 s / page | 0.5 s / page |
| Best for | Word-level OCR with positions | Lightweight OCR pipeline on CPU |
| Inputs | PDF, images | PDF, images |
| Languages | Mainly Latin scripts | Chinese and English |
| Max pages per request | 500 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | No | No |
| Multilingual | No | No |
| 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 docTR if
- You need a nested page, block, line and word structure with coordinates exported as JSON.
- You want to choose among several detection and recognition architectures, with straight or rotated boxes.
- You want ready Docker images and a Hugging Face Spaces demo, plus a key information extraction predictor under Apache-2.0.
Choose OpenOCR if
- Your documents are Chinese or English academic PDFs with formulas, and you want LaTeX output.
- You want lightweight document parsing: OpenDoc-0.1B reports 90.57 on OmniDocBench v1.5 with 0.1B parameters.
- You need ONNX export for deployment, or you do scene text recognition research.
Key differences
| docTR | OpenOCR | |
|---|---|---|
| Scope | Text detection and recognition. Returns words and lines only, with no table, formula or Markdown output. | Text detection, recognition, formula, table and document parsing. |
| Output formats | JSON, plain text, bounding boxes | Plain text, LaTeX, bounding boxes |
| Licence and price | Apache-2.0, free, self-hosted (you pay for compute) | Apache-2.0, free, self-hosted (you pay for compute) |
| How you run it | Python library on PyTorch, with Docker images. GPU images need CUDA 12.2 or newer on the host. | Python library you self-host. Hardware requirements and throughput are not published. |
| Benchmark | Not published | OpenDoc-0.1B scores 90.57 on OmniDocBench v1.5, self-reported by the authors with no independent replication found. |
| Languages | Not published | OpenDoc and UniRec support only Chinese and English. |
| Known failure | Split-crop merging misjoins readings across repeated characters (issue #2229). Cropped images give poorer results (issue #2007). ONNX export of MASTER truncates decoding steps (issue #2224). | Document parsing depends on a separate layout model, PP-DocLayoutV2 from PaddlePaddle. |
Sources
What each one is
docTR
Open-source OCR library from Mindee with separate text detection and recognition models.
docTR detailsOpenOCR
Fudan's open toolkit for text detection and recognition, formulas, tables and document parsing.
OpenOCR 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": "mindee/doctr","fallbacks": ["fudan/openocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
docTR alternativesOpenOCR alternativesBest OCR models in 2026