Hi everyone !
I’m currently making a scenario (whatsapp assistant) and it works perfectly, except for one important detail: it is very slow, as each Open AI API call takes 5-10 seconds, and as I have between 1 and 3 depending on the execution, the whole scenario takes around 30 sec which isn’t very dynamic…
Anyone knows if I can change this ? Ideally I would like the scenario to run much faster…
I would be glad to show more details about the scenario, and pay whoever can fix my issue 
Have a great day !
Hello @Nathan_Porterie, welcome to the make Hire a pro section , I would love to work with you on this, you can book a call Here to discuss the project requirements and you can checkout my upwork profile Here
Hi @Nathan_Porterie,
Great to see your WhatsApp assistant project coming along!
I understand the delay with your OpenAI API calls. To speed things up, the first step would be to reduce prompt length, which will help decrease processing time. Additionally, I’d consider batching multiple API calls to cut down on wait times.
I’d be happy to review your setup and provide a tailored solution. If you need this sorted quickly, I’m available on weekends too.
You can check my services, connect on LinkedIn or book a quick call to get started.
I look forward to your response and the opportunity to assist you further.
What’s the business case?
What’s the industry you are applying this to?
Without looking at it with more detail, 2 things comes to my mind:
- Merge multiple prompts into 1
- Use faster, less accurate models
But these are possible only testing and in specific contexts