RapidOCR vs InternVL3.5
Pick RapidOCR for offline CPU OCR of Chinese and English screenshots and receipts, or for embedding OCR in Java or C# apps. Pick InternVL3.5 for document question answering and Markdown or JSON output, if you have GPUs. The main split is a small text-and-box OCR library versus a self-hosted vision language model family.
RapidOCRrapidai/rapidocr | InternVL3.5opengvlab/internvl | |
|---|---|---|
| Type | Open source | Vision model |
| Price | 1 credit / page | 2 credits / page |
| Per 1,000 pages | $1.00 | $2.00 |
| Typical speed | 0.6 s / page | 3.0 s / page |
| Best for | Lightweight OCR on CPU or edge | Self-hosted document understanding |
| Inputs | images | images |
| Languages | Chinese, English and more | Multilingual |
| Max pages per request | 100 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | No |
| Confidence | ✓ Yes | 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 RapidOCR if
- It runs on CPU through ONNX Runtime by default, so you need no GPU.
- It has bindings for Python, C++, Java and C#, so it embeds in many stacks.
- It returns plain text and bounding boxes, which you need when a job wants word positions.
Choose InternVL3.5 if
- It outputs Markdown and JSON as well as plain text, and suits document question answering.
- It comes in nine sizes from 1B to 241B, so you can trade accuracy for cost.
- Models up to 30B fit on a single A100, and the MoE variants 20B-A4B and 30B-A3B activate few parameters per token.
Key differences
| RapidOCR | InternVL3.5 | |
|---|---|---|
| What it is | OCR toolkit that converts PaddleOCR models to ONNX and other runtimes | Open-weight vision language model family from OpenGVLab, 1B to 241B-A28B parameters |
| Licence | Apache-2.0; converted models keep their upstream licences | Apache-2.0 |
| How you run it | Library for Python, C++, Java or C#; pip install rapidocr onnxruntime | Self-host with Transformers, LMDeploy or vLLM |
| Hardware | CPU via ONNX Runtime by default; optional GPU, TensorRT, OpenVINO, MNN, Paddle or PyTorch engines | Up to 30B on a single A100; 38B needs two A100s; 241B-A28B needs eight A100s |
| Output | Plain text and bounding boxes | Plain text, Markdown and JSON; no bounding boxes or layout output by default |
| Languages | Chinese and English by default; other languages need extra models | Not published |
| Known failure | 16-bit grayscale PNG or TIFF give no text or wrong text; Arabic recognition crashes; sliced inference crashes on large pages | OCR scores are published only as chart images, not text values |
Sources
What each one is
RapidOCR
PaddleOCR models packaged for ONNX Runtime and other engines. Easy to deploy without Paddle.
RapidOCR 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": "rapidai/rapidocr","fallbacks": ["opengvlab/internvl"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
RapidOCR alternativesInternVL3.5 alternativesBest OCR models in 2026