Llama Parse Slowness and Variation in Output Using Parsing instruction

[qasim-mobi]

Describe the bug
Llama Parse gets extremely slowed down. It is taking several minutes even for returning the output and even took few minutes even for cached response and also another problem is that while using the accurate mode with same parsing instructions used earlier, Llama Parse is providing different output rather than which it was giving previously on same file.

Client:
Please remove untested options:

• Python Library

Additional context
Add any additional context about the problem here.
What options did you use?

• parsing_instruction
• donot_unroll_column
• accurate mode As well as vendor_multimodal_model

[hamzanouali42]

We are facing the same issue