Marker vs InternVL3.5
Pick Marker when you need PDFs converted to Markdown, HTML or JSON on your own machine, including CPU-only or Apple Silicon, and your company is under the weight-licence limits. Pick InternVL3.5 when you want a general vision language model under Apache-2.0 for document question answering, at a size you choose from 1B to 241B-A28B. The main split is purpose and licence: Marker is a PDF converter with restricted weights, InternVL3.5 is an open model with no commercial limit.
Markerdatalab/marker | InternVL3.5opengvlab/internvl | |
|---|---|---|
| Type | OCR API | Vision model |
| Price | 14 credits / page | 2 credits / page |
| Per 1,000 pages | $14.00 | $2.00 |
| Typical speed | 2.5 s / page | 3.0 s / page |
| Best for | Books and research papers | Self-hosted document understanding |
| Inputs | PDF, images, DOCX | images |
| Languages | 90+ languages | Multilingual |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| 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 Marker if
- You want a ready PDF to Markdown converter that also outputs HTML, JSON block trees, RAG chunks and LaTeX.
- You have no GPU: fast mode runs on CPU with about 18 MB of ONNX weights at 23.6 pages per second, though olmOCR-bench drops to 66.6.
- You prefer a hosted option: Datalab sells an API with about 5,000 free pages per month.
Choose InternVL3.5 if
- You are a company above $5M and need free commercial use, since the code and weights are Apache-2.0.
- You want to trade accuracy for cost across nine sizes from 1B to 241B-A28B, including MoE variants 20B-A4B and 30B-A3B.
- You need document question answering or a multimodal pipeline, not only conversion to Markdown.
Key differences
| Marker | InternVL3.5 | |
|---|---|---|
| What it is | PDF to Markdown converter built on the Surya OCR models, from Datalab | Open-weight vision language model family from OpenGVLab |
| Licence | Code Apache-2.0; weights under a modified AI Pubs Open Rail-M licence, free for research, personal use and startups under $5M | Apache-2.0 |
| How you run it | Locally on GPU, CPU or Apple Silicon, or through Datalab's hosted API | Self-host only, with Transformers, LMDeploy or vLLM |
| Hardware | vllm mode needs an NVIDIA GPU; fast mode runs on CPU | Up to 30B fits on a single A100; 38B needs two A100s; 241B-A28B needs eight A100s |
| Output formats | Markdown, HTML, JSON, chunks, LaTeX, images | Plain text, Markdown, JSON; no bounding boxes or layout output by default |
| Benchmark | olmOCR-bench 83.7 in vllm mode and 66.6 in CPU fast mode, all reported by Datalab | Not published as text; the model card shows OCR scores only as chart images |
| Price | Free self-hosted within the weight licence; hosted API has about 5,000 free pages per month, paid plans on Datalab's pricing page | Free self-hosted; you pay for compute |
Sources
What each one is
Marker
Hosted version of the open-source Marker and Surya models. Strong on books and papers.
Marker detailsInternVL3.5
Shanghai AI Lab's open vision-language family (1B to 241B, August 2025), strong on OCR and document benchmarks.
InternVL3.5 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": "datalab/marker","fallbacks": ["opengvlab/internvl"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Marker alternativesInternVL3.5 alternativesBest OCR models in 2026