Show HN: CPU-only fast OCR for screenshots, images, PDFs, webpages
6 points
14 hours ago
| 2 comments
| github.com
| HN
▲
explosion-s
10 hours ago
[-]
Why is no GPU a pull here? Is there usually an activation overhead GPU request or something?
reply
▲
mrkn1
10 hours ago
[-]
No, simply, my laptop only has a CPU.
reply
▲
arnabdey0503
14 hours ago
[-]
Have you done benchmark againest other methods?
reply
▲
mrkn1
10 hours ago
[-]
I haven't. But the evals of the underlying model are published here, including on Omnibench. https://huggingface.co/PaddlePaddle/PaddleOCR-VL-1.5
reply