`TextMonkey` model

In my quick tests on the demo, it seems to be the best document understanding and OCR model I have ever tested, my current use case is that I have to identify the process code of 1500000 images manually (a challenging job) (I am wondering if this model will be able to do this for me) I have to identify from an image like the one below what the process code/year is in each image  [573.pdf](https://github.com/user-attachments/files/15859661/573.pdf) https://github.com/Yuliang-Liu/Monkey?tab=readme-ov-file [TextMonkey](https://arxiv.org/abs/2403.04473)