olmOCR vs Typhoon OCR
Pick olmOCR for large-scale PDF to Markdown, academic papers with equations, and mixed printed and handwritten pages, when you have a GPU with at least 12 GB VRAM. Pick Typhoon OCR for Thai and English documents such as government forms, financial reports and books. Its main edge is Thai accuracy at only 2B parameters, and it also runs through a hosted API.
olmOCRallenai/olmocr | Typhoon OCRtyphoon/typhoon-ocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 1.5 s / page | 1.8 s / page |
| Best for | Bulk PDF to text for datasets | Thai documents |
| Inputs | PDF, images | PDF, images |
| Languages | Mainly English | Thai, English |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| Bounding boxes | No | No |
| Confidence | No | 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 olmOCR if
- Self-hosting costs under $200 per million pages per Ai2, and hosted olmOCR-2-7B on DeepInfra is $0.09 input / $0.19 output per million tokens.
- It handles equations, tables and handwriting, and strips headers and footers in natural reading order, which suits academic papers and RAG corpora.
- Weights, training data and code are all released, and the toolkit is Apache-2.0 with no extra terms.
Choose Typhoon OCR if
- It is built for Thai and English. V1.5 beat GPT-5 and Gemini 2.5 Pro on Thai government forms (BLEU 0.870), and V2 2B scores 81.3 on MDPBench Thai pages versus 61.3 for its base model.
- It outputs tables as HTML with merged cells, equations as LaTeX and marks checkboxes, which fits forms and financial reports.
- At 2B parameters (with a 0.8B V2 variant) it scores 84.7 on olmOCR-bench, so it needs far less model than olmOCR's 7B.
Key differences
| olmOCR | Typhoon OCR | |
|---|---|---|
| Licence | Apache-2.0 (toolkit) | Apache-2.0, subject to OpenTyphoon Terms and Conditions |
| Model size | 7B | 2B (V2 and V1.5); 0.8B V2 variant |
| How you run it | Self-host with vLLM on an NVIDIA GPU with at least 12 GB VRAM, or use third-party hosted endpoints | Self-host the weights, or use the Typhoon API as typhoon-ocr-v2 |
| Price | Under $200 per million pages self-hosted; DeepInfra $0.09 input / $0.19 output per million tokens | Free self-hosted (you pay for compute); Typhoon API pricing not published |
| Output formats | Markdown, JSON | Markdown, HTML, LaTeX |
| olmOCR-bench | 82.4 overall (v0.4.0) | 84.7 (V2 2B) |
| Languages | Not published | Only Thai and English evaluated |
| Known failure | Out of memory on a 12 GB GPU (issue #424); tends to get stuck on the last few pages (issue #433); unreliable image bounding boxes (issue #437) | V2 can loop until the token limit, mostly on dense multi-column pages; key-information extraction is weakest at 21.0 on ThaiOCRBench |
Sources
- olmOCR GitHub README
- Issue #424 OOM
- Issue #433 stuck
- Typhoon OCR V2 2B model card
- Typhoon OCR technical report
What each one is
olmOCR
Open-source vision model from Ai2 (often searched as olm ocr) that turns PDFs into clean, reading-order text. Latest release: olmOCR 2.
olmOCR detailsTyphoon OCR
Open vision model tuned for Thai and English documents, with Markdown output.
Typhoon OCR 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": "allenai/olmocr","fallbacks": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
olmOCR alternativesTyphoon OCR alternativesBest OCR models in 2026