Model request timed out

We keep getting multiple messages with model request timed out, and responding to the end-user takes 10 seconds. We're using gpt-4o + openai voice + deepgram. The assistant is created via make.com. I deleted all the functions from make. I set number of words to interrupt as 1.

Example call id: 71df6040-0752-4cf2-aa94-24d106a9b6b5
b4c56242-bc66-4c81-94dc-9cc2f8296f2b

org id: 08391a09-ead2-4bf7-975a-300f098158be

Our prompt is quite long so not sure if that affects that.

What can be causing this issue? It's quite urgent as all our customers hang up with this massive latency
Screenshot_2024-06-17_at_5.28.06_PM.png
Was this page helpful?