LlamaParse vs Typhoon OCR
Pick Typhoon OCR for Thai and Thai-English documents such as government forms, financial reports and books, especially if you want to self-host under Apache-2.0. Pick LlamaParse for mixed PDF and office files going into a RAG pipeline, where a hosted API with a free monthly allowance is the main draw. Neither should be trusted unchecked on structured fields: LlamaParse has invented bank statement transactions, and Typhoon OCR scores 21.0 on key-information extraction.
LlamaParsellamaindex/llamaparse | Typhoon OCRtyphoon/typhoon-ocr | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 5 credits / page | 2 credits / page |
| Per 1,000 pages | $5.00 | $2.00 |
| Typical speed | 3.0 s / page | 1.8 s / page |
| Best for | RAG ingestion | Thai documents |
| Inputs | PDF, images, DOCX | PDF, images |
| Languages | 80+ languages | Thai, English |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | No |
| 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 LlamaParse if
- You feed mixed PDF and office documents into a RAG pipeline, especially inside LlamaIndex.
- You want one hosted API with tiers, from 1 credit per page in Fast mode to 45 credits in Agentic Plus.
- You want to start free: 10K credits per month, and re-parsing the same file within 48 hours costs nothing.
Choose Typhoon OCR if
- Your documents are Thai or Thai-English, such as government forms, financial reports or books.
- You want open weights you can self-host, with a 2B model or a 0.8B V2 variant, and no per-page fee.
- You need tables as HTML with merged cells, equations as LaTeX and checkboxes marked.
Key differences
| LlamaParse | Typhoon OCR | |
|---|---|---|
| Licence | Proprietary | Apache-2.0, subject to OpenTyphoon Terms and Conditions |
| How you run it | Hosted API only | Self-host the weights, or use the Typhoon API (typhoon-ocr-v2) |
| Price | $1.25 per 1,000 credits; about $1.25 to $56 per 1,000 pages depending on tier; 10K free credits per month | Free to self-host (you pay for compute); API pricing not published on the model card |
| Output formats | Markdown, JSON, plain text, images | Markdown, HTML, LaTeX |
| Benchmarks | Not published | Author-reported: 84.7 olmOCR-bench, 90.97 OmniDocBench v1.6, 59.9 ThaiOCRBench (V2 2B) |
| Language scope | Not published | Thai and English only; no other languages evaluated |
| Known failure | Invented transactions on bank statements (#450); hallucination on low-quality scans with no confidence score (#483); jobs stuck for days and premium mode taking 4 to 5 minutes (#623) | V2 can loop until the token limit on dense multi-column pages; can output text not in the image; GGUF builds may lose accuracy |
Sources
- LlamaIndex pricing
- GitHub issue #450
- GitHub issue #483
- Typhoon OCR V2 2B model card
- Typhoon OCR technical report
What each one is
Typhoon 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": "llamaindex/llamaparse","fallbacks": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
LlamaParse alternativesTyphoon OCR alternativesBest OCR models in 2026