RolmOCR vs Chunkr
Pick RolmOCR if you want free, Apache-2.0 plain text transcription of scanned pages on your own GPU. Pick Chunkr if you need layout-aware output (HTML, Markdown, JSON with bounding boxes) for RAG chunking of PDFs and slides. The main split is output: RolmOCR returns plain text only, while Chunkr returns structure and coordinates.
RolmOCRreducto/rolmocr | Chunkrchunkr/chunkr | |
|---|---|---|
| Type | Open source | OCR API |
| Price | 2 credits / page | 14 credits / page |
| Per 1,000 pages | $2.00 | $14.00 |
| Typical speed | 3.0 s / page | 3.0 s / page |
| Best for | Fast open PDF to text | Layout-aware chunks for RAG |
| Inputs | PDF, images | PDF, images, DOCX |
| Languages | Not stated by the authors | Multilingual |
| Max pages per request | 500 | 1,000 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ 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 RolmOCR if
- You only need plain text from scanned pages, including rotated or skewed ones, since about 15% of its training data was rotated.
- You want Apache-2.0 weights that are usable commercially, with no copyleft or commercial license to negotiate.
- You want a single self-hosted model on one GPU, with no extra LLM endpoint to run.
Choose Chunkr if
- You need bounding boxes, which RolmOCR cannot output according to its model card.
- You need HTML, Markdown or JSON output and semantic chunking for RAG over PDFs, PPT, Word and images.
- You want the choice of a hosted cloud API or a self-host via Docker with GPU, CPU-only and Mac ARM options.
Key differences
| RolmOCR | Chunkr | |
|---|---|---|
| What it is | Open-weight OCR model from Reducto, fine-tuned from Qwen2.5-VL-7B-Instruct (8B params) | Document parsing service from Lumina AI: layout analysis, OCR with bounding boxes, semantic chunking |
| Output formats | Plain text | HTML, Markdown, JSON, bounding boxes |
| Deployment | Self-host only, for example with vLLM | Cloud API or self-host via Docker (GPU, CPU-only, Mac ARM) |
| Licence | Apache-2.0, usable commercially | AGPL-3.0 or commercial license; closed-source use needs the commercial license |
| Price | Free, you pay for compute | Hosted credit-based plans, official prices not verified. Open-source version is free, you pay for compute and an LLM endpoint |
| Input types | Scanned pages, including rotated or skewed ones | PDF, PPT, Word and images; Excel is not supported in the open-source version |
| Known failure | Model card says it can hallucinate or drop content. Reducto shows it omitting a table subtitle. Needl.ai overall score 0.663 | Open-source version uses community models that are less accurate than the cloud's in-house models. Benchmarks: Not published |
Sources
What each one is
RolmOCR
Reducto's open PDF-to-text model, built on Qwen2.5-VL-7B with the olmOCR recipe. Needs no PDF metadata and handles rotated pages.
RolmOCR 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": "reducto/rolmocr","fallbacks": ["chunkr/chunkr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
RolmOCR alternativesChunkr alternativesBest OCR models in 2026