Tesseract vs Chandra
Pick Tesseract for clean printed scans, offline searchable PDFs and cheap CPU batch jobs in many languages. Pick Chandra when you need Markdown, HTML or JSON with layout from scans that contain tables, forms or math, and you can run a GPU or pay for the Datalab API. The main split is plain text lines on a CPU versus structured document output from a vision-language model.
Tesseracttesseract/tesseract | Chandradatalab/chandra | |
|---|---|---|
| Type | Open source | Open source |
| Price | Free | 2 credits / page |
| Per 1,000 pages | $0.00 | $2.00 |
| Typical speed | 0.6 s / page | 2.0 s / page |
| Best for | Clean printed text, free | Complex layouts and handwriting |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | 40+ languages |
| Max pages per request | 500 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | ✓ Yes |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | No |
| Forms | No | ✓ Yes |
| 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
- It runs on CPU only with no model server, so it fits small machines and offline use.
- It is Apache-2.0 with no revenue limit, and it recognizes more than 100 languages out of the box.
- It outputs hOCR, searchable PDF, TSV, ALTO and PAGE XML, and it is the engine behind OCRmyPDF.
Choose Chandra if
- It returns Markdown, HTML or JSON with bounding boxes and block labels, which suits PDF to Markdown for RAG pipelines.
- Chandra 2 scores 85.8 on olmOCR-bench and averages 77.8% across 43 languages, per the authors' tables.
- It is aimed at documents with tables, forms and math, where Tesseract returns only lines and words.
Key differences
| Tesseract | Chandra | |
|---|---|---|
| Licence | Apache-2.0, no usage limits | Code Apache-2.0; weights modified OpenRAIL-M, free for research, personal use and companies under $2M funding or revenue |
| How you run it | Local CLI or C++ library, CPU only | Self-host on a single NVIDIA GPU via vLLM, or use the hosted Datalab API |
| Price | Free, you pay for compute | Self-hosted free within licence terms; API is $1, $3 or $7.50 per 1,000 pages, with $20/month free tier |
| Output | Plain text, hOCR, searchable PDF, TSV, ALTO, PAGE; no tables, formulas or reading order | Markdown, HTML, JSON and bounding boxes |
| Benchmarks | Not published | 85.8 on olmOCR-bench (Chandra 2, authors' figure) |
| Input quality | Needs binarization, deskew and about 300 DPI; weak on skewed or noisy photos | Old Scans category scores only 51.1 on olmOCR-bench |
| Known failure | Confuses digit 1 with lowercase l (issue #2753); not built for handwriting | Token-loop hallucinations and missing pages on some Russian and Kazakh scans (issue #51); 43.2% on Urdu |
Sources
- Tesseract GitHub README
- Improving quality (tessdoc)
- Issue #2753 1 vs l
- Chandra GitHub README
- Datalab pricing
- GitHub issue #51 token loops
What each one is
Chandra
Datalab's OCR model that keeps layout and handles forms, tables and handwriting.
Chandra 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": ["datalab/chandra"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Tesseract vs SuryaTesseract vs olmOCRPaddleOCR (PP-OCR) vs ChandraDocling vs ChandraAmazon Textract vs TesseractPaddleOCR-VL vs TesseractDocling vs TesseractEasyOCR vs Chandra
Tesseract alternativesChandra alternativesBest OCR models in 2026