docTR vs Chunkr
Pick docTR if you want a Python OCR library that returns words and lines with coordinates, runs locally, and carries an Apache-2.0 licence. Pick Chunkr if you need documents turned into HTML or Markdown chunks, for example to feed a RAG pipeline from PDFs and slides. The main split is output: docTR stops at words and lines, while Chunkr does layout parsing and chunking.
docTRmindee/doctr | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 1 credit / page | 14 credits / page |
| Per 1,000 pages | $1.00 | $14.00 |
| Typical speed | 0.9 s / page | 3.0 s / page |
| Best for | Word-level OCR with positions | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | Mainly Latin scripts | Multilingual |
| Max pages per request | 500 | 1,000 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | 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 docTR if
- You want word-level OCR with a nested page, block, line and word structure and coordinates exported as JSON.
- You need Apache-2.0 licensing, with no commercial licence needed for closed-source products.
- You have a Python ML team that wants to choose detection and recognition architectures, with straight or rotated boxes, and train its own models.
Choose Chunkr if
- You need Markdown or HTML output from PDF, PPT, Word or image files, which docTR does not produce.
- You are building RAG over PDFs and slides and want layout analysis and semantic chunking.
- You want a choice between a hosted cloud API and a self-hosted Docker setup with GPU, CPU-only or Mac ARM options.
Key differences
| docTR | Chunkr | |
|---|---|---|
| Licence | Apache-2.0 | AGPL-3.0 or commercial licence; closed-source use needs the commercial licence |
| How you run it | Python library on PyTorch, with Docker images | Cloud API or self-hosted open-source version via Docker |
| Price | Free, self-hosted; you pay for compute | Hosted credit-based plans (prices not verified); open-source version is free, you pay for compute and an LLM endpoint |
| Output formats | JSON, plain text, bounding boxes | HTML, Markdown, JSON, bounding boxes |
| Tables and Markdown | No table, formula or Markdown output; words and lines only | Markdown and HTML output with layout analysis |
| Hardware and dependencies | CPU or NVIDIA GPU; GPU Docker images need CUDA 12.2 or newer on the host | Self-host needs an external OpenAI-compatible LLM endpoint |
| Known issues | Split-crop merging can misjoin readings across repeated characters (issue #2229); cropped images give poorer results (issue #2007); ONNX export of MASTER truncates decoding steps (issue #2224) | Open-source version uses community models that are less accurate than the cloud's in-house models; Excel is not supported in the open-source version |
Sources
What each one is
docTR
Open-source OCR library from Mindee with separate text detection and recognition models.
docTR detailsChunkr
Document parsing API that segments and chunks pages for RAG. Self-hostable version available.
Chunkr 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": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
docTR alternativesChunkr alternativesBest OCR models in 2026