SmolDocling vs Llama 3.2 90B Vision
SmolDocling is cheaper per page. SmolDocling is faster. Only SmolDocling supports Tables, Math, Bounding boxes. Only Llama 3.2 90B Vision supports Multilingual.
SmolDoclinghuggingface/smoldocling | Llama 3.2 90B Visionmeta/llama-3-2-90b-vision | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 3 credits / page |
| Per 1,000 pages | $1.00 | $3.00 |
| Typical speed | 1.0 s / page | 3.0 s / page |
| Best for | Very cheap page conversion | Self-hosted image questions |
| Inputs | PDF, images | images |
| Languages | Mainly English | English for vision |
| Max pages per request | 500 | 50 |
| Tables | ✓ Yes | No |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | No |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | No | ✓ Yes |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
What each one is
SmolDocling
Tiny 256M vision model from Hugging Face and IBM that outputs DocTags. Predecessor of Granite Docling.
SmolDocling detailsLlama 3.2 90B Vision
Meta's larger Llama 3.2 vision model (September 2024).
Llama 3.2 90B Vision detailsTry both on your own file
Run the file once with each model and compare the output.
PDF, PNG, JPG, WebP, HEIC. Up to 10 files at onceNo file?
Use one, fall back to the other
{"model": "huggingface/smoldocling","fallbacks": ["meta/llama-3-2-90b-vision"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
SmolDocling vs Llama 3.2 11B VisionSmolDocling vs Llama 4 ScoutMonkeyOCR vs Llama 3.2 90B VisionGranite Docling vs Llama 3.2 90B VisionClaude vs SmolDoclingGPT vs SmolDoclingdots.ocr vs SmolDoclingPaddleOCR-VL vs SmolDocling
SmolDocling alternativesLlama 3.2 90B Vision alternativesBest OCR models in 2026