Zoho Deluge API call to Open AI time-out issue

We are looking for Zoho Deluge experts with experience in integration with Open AI. Do you have such a specific experience? Can you please share your portfolio?

We are developing an integration between Zoho CRM and Open AI and we currently have an issue we would like to solve, and we are also looking for a long-term partner.

Can you please quote for a solution to the following issue and provide your portfolio of experience of intergration Zoho Deluge/Open AI:

Please, find below the code: the question is "How can I overcome the 40-second timeout issue within Zoho without having to break down the Open AI prompt/output into smaller chunks and thus finding a solution for longer responses within the token limits of gpt turbo 3.5?". In particular, I would like to get longer outputs (e.g. more than 3 improvement points and longer than 200 characters, which is the current output it seems to be able to manage in 40 seconds). See prompt in the code attached.

We understand that the OpenAI API call allows to stream chunks of response from the API call: https://github.com/openai/openai-cookbook/blob/main/examples/How_to_stream_completions.ipynb , but can't find a way to "flush" the chunks out of the response in the invokeURL Zoho Deluge method to allow at least to recover the chunks that were generated within the 40-seconds time-out window you guys impose - even if we believe a better solution would be to launch a call for longer outputs and then close it within the 40-seconds and get an handler to go back pick it up a few moments later when completed.

I look forward to receiving your quotation and portfolio.

The sample code is attached.