OCRRouter

Nougat vs InternVL3.5

Pick Nougat only for non-commercial work on English scientific papers, where you need LaTeX for equations and tables. Pick InternVL3.5 for commercial use, other document types, or when you want to choose a model size. The main split is licence: Nougat weights are CC-BY-NC, InternVL3.5 is Apache-2.0.

Nougatmeta/nougatInternVL3.5opengvlab/internvl
TypeOpen sourceVision model
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed3.0 s / page3.0 s / page
Best forarXiv-style papersSelf-hosted document understanding
InputsPDF, imagesimages
LanguagesMainly EnglishMultilingual
Max pages per request200500
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxesNoNo
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 Nougat if

  • It was trained on arXiv and PubMed Central papers, so it targets academic PDFs directly.
  • It outputs Mathpix-compatible Markdown with LaTeX for equations and tables.
  • It ships small and base checkpoints with a simple CLI and API.

Choose InternVL3.5 if

  • Apache-2.0 licence allows commercial use, while Nougat weights are CC-BY-NC.
  • Nine sizes from 1B to 241B let you trade accuracy for cost, and models up to 30B fit on a single A100.
  • It is a general vision language model, so it suits document question answering and self-hosted multimodal pipelines.

Key differences

NougatInternVL3.5
LicenceCode MIT; weights CC-BY-NC (no commercial use)Apache-2.0
How you run itSelf-hosted Python package; free for non-commercial use, you pay for computeSelf-hosted with Transformers, LMDeploy or vLLM; free, you pay for compute
Model sizesSmall and base checkpointsNine sizes, 1B to 241B-A28B; 8B variant is 8.5B
Output formatsMarkdown and LaTeXPlain text, Markdown and JSON; no bounding boxes or layout output by default
Best document typesEnglish arXiv-style scientific papers, math-heavy PDFsDocument question answering and multimodal pipelines
LanguagesWorks best on English; README says Chinese, Russian and Japanese do not workNot published
Known failureRepetition loops on 1.5% of test pages in the paper, more on out-of-domain documents; repetition detection can drop pages as MISSING_PAGEOCR scores are published only as chart images, not text values

Sources

What each one is

Nougat

Meta's model that converts academic PDFs to Markdown with LaTeX math.

Nougat 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": "meta/nougat","fallbacks": ["opengvlab/internvl"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Nougat alternativesInternVL3.5 alternativesBest OCR models in 2026