Spaces:

---

prithivMLmods

/

core-OCR

like 220

Running on Zero

I gave this image (one part only) sudoku game to fill the numbers. but its failed

@Senthilds08
You’re trying to apply logical problems to the OCR models basically, they’re aligned for content understanding and content extraction (OCR) on documents/entries.
Try the vision-language multimodal models that excel in logical reasoning and problem-solving alignments.