InternVL3.5 vs OCR.space
Pick InternVL3.5 if you want to self-host an Apache-2.0 vision language model for document question answering and can supply GPUs. Pick OCR.space if you want a hosted API with a free tier, word coordinates and searchable PDFs, and do not want to run any infrastructure. The main split is who runs the model: you run InternVL3.5, OCR.space runs itself.
InternVL3.5opengvlab/internvl | OCR.spaceocrspace/ocr-space | |
|---|---|---|
| Type | Vision model | OCR API |
| Price | 2 credits / page | 1 credit / page |
| Per 1,000 pages | $2.00 | $1.00 |
| Typical speed | 3.0 s / page | 1.5 s / page |
| Best for | Self-hosted document understanding | Cheap simple OCR |
| Inputs | images | PDF, images |
| Languages | Multilingual | 20+ languages |
| Max pages per request | 500 | 100 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | No | ✓ Yes |
| 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 InternVL3.5 if
- The Apache-2.0 licence and free weights mean you pay only for compute, with no per-request fee.
- Nine sizes from 1B to 241B-A28B let you trade accuracy against cost, and models up to 30B fit on a single A100.
- It answers questions about a document and can return plain text, Markdown or JSON, so it suits self-hosted multimodal pipelines.
Choose OCR.space if
- You call a hosted API, so there are no GPUs to provision, and the free tier gives 25,000 requests per month.
- It returns per-word overlay coordinates, bounding boxes and searchable PDFs, which InternVL3.5 does not give by default.
- Engine 3 covers 200+ languages, tables, handwriting and checkboxes, and PRO plans carry a 100% uptime or money back SLA.
Key differences
| InternVL3.5 | OCR.space | |
|---|---|---|
| How you run it | Self-host with Transformers, LMDeploy or vLLM | Hosted API; on-premise version sold separately |
| Licence | Apache-2.0, open weights | Proprietary |
| Price | Free software; you pay for compute | Free: 25,000 requests/month. PRO $30/month for 300,000 requests. PRO PDF $60/month. Enterprise from $999/month |
| Hardware or limits | Up to 30B on one A100; 38B needs two A100s; 241B-A28B needs eight A100s | Free tier: 1 MB files, 3 PDF pages. PRO: 5 MB. PRO PDF: 100 MB+, 999+ pages |
| Output formats | Plain text, Markdown, JSON | JSON, plain text, bounding boxes, searchable PDF, Markdown |
| Word coordinates and layout | No bounding boxes or layout output by default | Per-word overlay coordinates, but Engine 3 coordinates are less precise and overlay requests are 2 to 3 times slower |
| Published OCR benchmark | Not published as text; the model card shows OCR scores only as chart images | Not published |
Sources
What each one is
InternVL3.5
Shanghai AI Lab's open vision-language family (1B to 241B, August 2025), strong on OCR and document benchmarks.
InternVL3.5 detailsOCR.space
Low-cost OCR API with a large free tier. Returns text, word boxes and searchable PDFs.
OCR.space 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": "opengvlab/internvl","fallbacks": ["ocrspace/ocr-space"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
InternVL3.5 alternativesOCR.space alternativesBest OCR models in 2026