DeepSeek-OCR vs Chunkr
Pick DeepSeek-OCR if you want a free, MIT-licensed model you run yourself for cheap bulk OCR on a CUDA GPU. Pick Chunkr if you need layout parsing and chunking of PDFs, slides, Word files and images for RAG, with a hosted API or a self-host option. The main split is a single OCR model versus a document parsing service with structured output.
DeepSeek-OCRdeepseek/deepseek-ocr | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 1 credit / page | 14 credits / page |
| Per 1,000 pages | $1.00 | $14.00 |
| Typical speed | 1.1 s / page | 3.0 s / page |
| Best for | Cheap bulk conversion | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | 100+ languages | Multilingual |
| Max pages per request | 200 | 1,000 |
| 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 DeepSeek-OCR if
- The MIT license allows commercial use with no copyleft terms, unlike Chunkr's AGPL-3.0.
- It offers resolution modes from 64 to 400 vision tokens per page, and the authors quote about 2,500 tokens per second of PDF throughput on one A100-40G.
- It runs with Transformers or vLLM and needs no external LLM endpoint, so cost is only your compute.
Choose Chunkr if
- It does layout analysis and semantic chunking, which suits RAG pipelines over PDFs and slides.
- It accepts PDF, PPT, Word and images, and returns HTML, Markdown, JSON and bounding boxes.
- It has a hosted cloud API, and the self-host version runs via Docker on GPU, CPU-only or Mac ARM.
Key differences
| DeepSeek-OCR | Chunkr | |
|---|---|---|
| Deployment | Self-host only, with Transformers or vLLM on a CUDA GPU | Cloud API or self-host via Docker (GPU, CPU-only, Mac ARM) |
| Licence | MIT | AGPL-3.0, or a commercial license for closed use |
| Price | Free, you pay for compute | Hosted credit-based plans (prices not published here); open-source version is free but you pay for compute and an LLM endpoint |
| Output formats | Markdown, plain text, HTML, bounding boxes | HTML, Markdown, JSON, bounding boxes |
| Input types | Page images and PDFs | PDF, PPT, Word and images; Excel is not supported in the open-source version |
| Extra dependencies | None beyond the model and a GPU | Self-host needs an external OpenAI-compatible LLM endpoint |
| Known failure | Repeated tokens and looping over table headers; it corrects nonsense words instead of copying them | Open-source version uses community models, less accurate than the cloud's in-house models |
Sources
- DeepSeek-OCR GitHub
- Visual Merit or Linguistic Crutch paper
- LearnOpenCV test
- Chunkr GitHub README
- Chunkr site
What each one is
Chunkr
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.
PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?
Use one, fall back to the other
{"model": "deepseek/deepseek-ocr","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
DeepSeek-OCR vs ABBYY FineReaderDeepSeek-OCR vs UnstructuredJina OCR vs ChunkrGoogle Document AI vs ChunkrLlamaParse vs ChunkrGoogle Document AI vs DeepSeek-OCRDeepSeek-OCR vs SuryaNougat vs Chunkr
DeepSeek-OCR alternativesChunkr alternativesBest OCR models in 2026