OCRRouter

OpenOCR vs InternVL3.5

Pick OpenOCR for self-hosted parsing of Chinese and English academic PDFs with formulas, where its 0.1B models keep compute low. Pick InternVL3.5 for document question answering or Markdown and JSON output, and when you want one model family you can size from 1B to 241B-A28B. The main split is that OpenOCR is a narrow OCR toolkit with bounding boxes and LaTeX, while InternVL3.5 is a general vision language model without coordinates by default.

OpenOCRfudan/openocrInternVL3.5opengvlab/internvl
TypeOpen sourceVision model
Price1 credit / page2 credits / page
Per 1,000 pages$1.00$2.00
Typical speed0.5 s / page3.0 s / page
Best forLightweight OCR pipeline on CPUSelf-hosted document understanding
InputsPDF, imagesimages
LanguagesChinese and EnglishMultilingual
Max pages per request500500
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxes✓ YesNo
ConfidenceNoNo
FormsNoNo
MultilingualNo✓ Yes
On OCRRouterAvailableAvailable

Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.

Which to choose

Choose OpenOCR if

  • OpenDoc-0.1B reports 90.57 on OmniDocBench v1.5 with only 0.1B parameters, though the authors self-report this score.
  • It returns bounding boxes and LaTeX, and UniRec-0.1B reads plain text and single or multi-line formulas in one model.
  • It supports ONNX export for deployment, and its OCR system reports 4.5% higher accuracy than PP-OCRv4 at similar speed.

Choose InternVL3.5 if

  • It comes in nine sizes from 1B to 241B-A28B, so you can trade accuracy for cost.
  • It outputs Markdown and JSON and suits document question answering, which OpenOCR does not list.
  • Models up to 30B run on a single A100, and the 20B-A4B and 30B-A3B MoE variants activate few parameters per token.

Key differences

OpenOCRInternVL3.5
LicenceApache-2.0Apache-2.0
Model typeOCR toolkit for detection, recognition, formula, table and document parsing, used as a Python libraryGeneral vision language model family, used through Transformers, LMDeploy or vLLM
Model size0.1B (OpenDoc-0.1B and UniRec-0.1B)1B to 241B-A28B; the 8B variant is 8.5B
Output formatsPlain text, LaTeX, bounding boxesPlain text, Markdown, JSON. No bounding boxes or layout output by default
LanguagesOpenDoc and UniRec support only Chinese and EnglishNot published
Benchmark90.57 on OmniDocBench v1.5 (OpenDoc-0.1B), self-reported by the authorsNot published as text. OCR scores appear only as chart images on the model card
HardwareNot published. No hardware requirements or throughput figuresUp to 30B on one A100, 38B on two A100s, 241B-A28B on eight A100s

Sources

What each one is

OpenOCR

Fudan's open toolkit for text detection and recognition, formulas, tables and document parsing.

OpenOCR details

InternVL3.5

Shanghai AI Lab's open vision-language family (1B to 241B, August 2025), strong on OCR and document benchmarks.

InternVL3.5 details

Try 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": "fudan/openocr","fallbacks": ["opengvlab/internvl"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

OpenOCR alternativesInternVL3.5 alternativesBest OCR models in 2026