Show HN: CPU-only fast OCR for screenshots, images, PDFs, webpages
github.comDo you have some benchmark on how fast it is on cpu? I checked the repo and the huggingface page but only found gpu-based benchmarks
I have not. My experience has been a few seconds for an 1024x1024 with medium density of text, FWIW. Feel free to try it on a few test images, model is pretty small and fast, but yeah no formal evals on CPU.
Why is no GPU a pull here? Is there usually an activation overhead GPU request or something?
No, simply, my laptop only has a CPU.
Have you done benchmark againest other methods?
I haven't. But the evals of the underlying model are published here, including on Omnibench. https://huggingface.co/PaddlePaddle/PaddleOCR-VL-1.5
Can it support other languages?
It should support 109 languages. More info here: https://huggingface.co/PaddlePaddle/PaddleOCR-VL