Granite Docling vs Chunkr
Pick Granite Docling if you run Docling on English PDFs full of tables, code and equations and want a 258M-parameter, Apache-2.0 model on cheap hardware. Pick Chunkr if you need a hosted API or a self-hosted service that parses PDFs, slides and Word files into chunks for RAG. The main split is a single document-conversion model versus a full parsing service.
Granite Doclingibm/granite-docling | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 1 credit / page | 14 credits / page |
| Per 1,000 pages | $1.00 | $14.00 |
| Typical speed | 1.5 s / page | 3.0 s / page |
| Best for | Small-model document conversion | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | English, experimental CJK and Arabic | Multilingual |
| Max pages per request | 500 | 1,000 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| 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.
Which to choose
Choose Granite Docling if
- You need Apache-2.0 licensing with no commercial-license step for closed products.
- Your documents are English PDFs with tables, code or equations, where the model card reports table structure TEDS 0.97 on FinTabNet and equation F1 0.968.
- You want low-cost conversion on CPU or a small GPU, since the model has only 258M parameters.
Choose Chunkr if
- You want a cloud API as well as a self-host option, instead of running everything yourself.
- Your inputs include PDF, PPT, Word and images, and you want semantic chunking and layout analysis for RAG.
- You already have an OpenAI-compatible LLM endpoint and want a Docker deployment with GPU, CPU-only or Mac ARM options.
Key differences
| Granite Docling | Chunkr | |
|---|---|---|
| Licence | Apache-2.0 | AGPL-3.0 or commercial license; closed-source products need the commercial license |
| How you run it | Self-hosted only, typically inside Docling | Cloud API or self-host via Docker; self-host needs an external LLM endpoint |
| Price | Free; you pay for compute | Hosted credit-based plans (official prices not verified); open-source version is free but you pay for compute and an LLM endpoint |
| Output formats | DocTags, Markdown, HTML, LaTeX, bounding boxes | HTML, Markdown, JSON, bounding boxes |
| Published benchmarks | Table TEDS 0.97 (FinTabNet), full-page OCR F1 0.84, equation F1 0.968, OCRBench 500, all from the model authors | Not published |
| Languages | English primary; Japanese, Arabic and Chinese are experimental | Not published |
| Known failure modes | Repetitive loops on page headings and invoices; exclamation-mark output on GPUs without bfloat16 such as T4 unless run in float32; vLLM wte error needs the untied revision | Open-source version uses community models that are less accurate than the cloud's; no Excel support in the open-source version |
Sources
- Granite-Docling model card
- HF discussion #24 repetition
- Ollama issue #12355
- Chunkr GitHub README
- Chunkr site
What each one is
Granite Docling
Compact IBM vision model that converts pages to DocTags for the Docling pipeline.
Granite Docling detailsChunkr
Document parsing API that segments and chunks pages for RAG. Self-hostable version available.
Chunkr 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": "ibm/granite-docling","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Granite Docling alternativesChunkr alternativesBest OCR models in 2026