Typhoon OCR vs Chunkr
Pick Typhoon OCR for Thai and Thai-English documents such as government forms, financial reports and books, where it has published scores. Pick Chunkr for RAG chunking of PDFs, slides and Word files, where you need layout analysis and bounding boxes. They solve different problems: Typhoon OCR is a 2B OCR model, Chunkr is a parsing service that needs an LLM endpoint when self-hosted.
Typhoon OCRtyphoon/typhoon-ocr | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 2 credits / page | 14 credits / page |
| Per 1,000 pages | $2.00 | $14.00 |
| Typical speed | 1.8 s / page | 3.0 s / page |
| Best for | Thai documents | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | Thai, English | Multilingual |
| Max pages per request | 500 | 1,000 |
| Tables | ✓ Yes | ✓ 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 Typhoon OCR if
- Your documents are Thai or Thai-English: V1.5 scored BLEU 0.870 on Thai government forms, ahead of GPT-5 and Gemini 2.5 Pro.
- You want permissive weights: Apache-2.0, subject to the OpenTyphoon Terms and Conditions, with a 2B model (and a 0.8B V2 variant) you run yourself.
- You need tables as HTML with merged cells, equations as LaTeX and checkboxes marked, with published scores of 84.7 on olmOCR-bench and 90.97 on OmniDocBench v1.6.
Choose Chunkr if
- You are building RAG over PDFs, PPT, Word files and images and want semantic chunking with bounding boxes.
- You want self-hosted layout parsing through Docker, with GPU, CPU-only and Mac ARM options.
- You already have an OpenAI-compatible LLM endpoint to plug in, or you want the hosted cloud API with credit-based plans.
Key differences
| Typhoon OCR | Chunkr | |
|---|---|---|
| What it is | Open 2B OCR model for Thai and English, also on the Typhoon API as typhoon-ocr-v2 | Document parsing service with layout analysis, OCR and semantic chunking |
| Licence | Apache-2.0, subject to OpenTyphoon Terms and Conditions | AGPL-3.0 or commercial licence; closed-source use needs the commercial licence |
| How you run it | Self-host the weights, or use the Typhoon API | Cloud API, or self-host via Docker; self-host needs an external LLM endpoint |
| Price | Free self-hosted (you pay for compute); API pricing not listed | Hosted credit-based plans, prices not verified; open-source version free (you pay for compute and an LLM endpoint) |
| Output formats | Markdown, HTML, LaTeX | HTML, Markdown, JSON, bounding boxes |
| Benchmarks | ThaiOCRBench 59.9, olmOCR-bench 84.7, OmniDocBench v1.6 90.97 (V2 2B, by the authors) | Not published |
| Known failure | V2 can loop until the token limit, mostly on dense multi-column pages; key-information extraction scores 21.0 on ThaiOCRBench | Open-source version uses community models, less accurate than the cloud's in-house models; no Excel support in open-source |
Sources
What each one is
Typhoon OCR
Open vision model tuned for Thai and English documents, with Markdown output.
Typhoon OCR 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": "typhoon/typhoon-ocr","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Typhoon OCR alternativesChunkr alternativesBest OCR models in 2026