docling-project / docling-project/docling
Enable force_ocr with other OpenAI compatible VLM providers
Open
enhancement
- Dominant language
- Python
- Stars
- 66.4k
- Forks
- 4.8k
- Avg merge
- 2d 21h
- Merged PRs (30d)
- 84
Description
### Requested feature
Right now we can use VLM to extract info from texts in images. But we can't do full page OCR except (EasyOcrOptions, TesseractOcrOptions, OcrMacOptions, RapidOcrOptions, TesseractCliOcrOptions)
I am using this doc as a reference: https://docling-project.github.io/docling/examples/full_page_ocr/
I would love to use OpenAI as a VLM for (FULL page OCR, i.e. force_ocr=true)
### Alternatives
I don't think any alternatives exist
Contributor guide
Assessment
This issue has not been assessed yet.