Tesseract vs Chunkr
Pick Tesseract for clean printed scans, offline searchable PDFs and cheap CPU batch OCR in many languages. Pick Chunkr when you need layout-aware output (HTML, Markdown, JSON with bounding boxes) from PDFs, slides and Word files, for example to chunk documents for RAG. The main split is plain lines and words from Tesseract versus structured, chunked parsing from Chunkr.
Tesseracttesseract/tesseract | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | Free | 14 credits / page |
| Per 1,000 pages | $0.00 | $14.00 |
| Typical speed | 0.6 s / page | 3.0 s / page |
| Best for | Clean printed text, free | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | 100+ languages | Multilingual |
| Max pages per request | 500 | 1,000 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ 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 Tesseract if
- You need OCR on CPU only, with no model server and no LLM endpoint, including on small or offline machines.
- You need more than 100 languages out of the box, or searchable PDF, hOCR, TSV, ALTO or PAGE XML output.
- You want an Apache-2.0 engine with no commercial licence question, and the engine behind OCRmyPDF.
Choose Chunkr if
- You need layout analysis and semantic chunking of PDFs, slides and Word files for RAG.
- You need HTML, Markdown or JSON output with bounding boxes, which Tesseract does not give (it returns lines and words only).
- You want a choice between a hosted cloud API and a Docker self-host with GPU, CPU-only or Mac ARM options.
Key differences
| Tesseract | Chunkr | |
|---|---|---|
| Licence | Apache-2.0 | AGPL-3.0 or commercial licence; closed-source use needs the commercial licence |
| How you run it | Local CLI or C++ library with an LSTM recognizer; CPU only, no GPU needed | Cloud API, or self-host via Docker on GPU, CPU-only or Mac ARM; self-host needs an external OpenAI-compatible LLM endpoint |
| Price | Free, self-hosted (you pay for compute) | Hosted credit-based plans, official prices not verified; open-source version is free (you pay for compute and an LLM endpoint) |
| Output formats | Plain text, hOCR, searchable PDF, TSV, ALTO, PAGE | HTML, Markdown, JSON, bounding boxes |
| Tables, formulas, reading order | No table structure, formula or reading-order output | Layout analysis and semantic chunking; table and formula support not published |
| Input handling | Needs binarization, deskew and about 300 DPI input for good accuracy; known 1 vs l confusion (issue #2753) | Accepts PDF, PPT, Word and images; Excel not supported in the open-source version |
| Handwriting and benchmarks | Not built for handwriting; no benchmark published | Handwriting not published; no benchmark published |
Sources
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": "tesseract/tesseract","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Tesseract vs ABBYY FineReaderTesseract vs UnstructuredPaddleOCR (PP-OCR) vs ChunkrDocling vs ChunkrAmazon Textract vs TesseractPaddleOCR-VL vs TesseractDocling vs TesseractFlorence-2 vs Chunkr
Tesseract alternativesChunkr alternativesBest OCR models in 2026