Did anyone try new gpt4 o model for text extraction from an image?

Did anyone try extracting data from images using latest gpt4 o model?

I was using gpt4 vision and accuracy was not that up to mark.
Do we see any improvement in newer model ? Or any ideas how to enhance the accuracy?

I Tried to use the 4-o model for generating a PlantUML code for a large flowchart. It seems to work better than gpt4vision for some reason.
Overall its working for me and found its results to better than the previous model.