I have set up an assistant with a configured knowledge base, and it is working well in terms of accurately answering user queries. However, I'm experiencing a noticeable lag during calls, particularly between user input and the assistant’s response. I attempted to reduce this by updating the assistant’s system prompt to include filler responses while processing, but the lag still persists. Could anyone please advise if there are any configurations, best practices, or optimizations available in Vapi to help minimize or eliminate this latency? Please let us know if you need any additional details from my side.

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards, Vapi Customer Support Team

Hey yes, this is a common Vapi issue and usually not the prompt. Most of the latency comes from LLM response time + knowledge base retrieval + TTS, not from silence handling. @Palak Jain

Hi Palak, could you please provide us the Call ID?

For call lag, you can refer this call id - 019c278c-5b7b-7ddb-8a1c-fcb8d25fb422 For example in this call recording you can check in the end when user said - "No. I don't have any other question." Then there is lag before assistant response - "Alright. Got it. So, hey, would you like me to connect you with 1 of our agents?

Hi Palak, Thanks for sharing the call ID. We’ve checked the logs for that call and don’t see any errors or abnormal behavior. The delay you’re noticing is expected in some cases and usually comes from a combination of LLM response time or TTS generation. Even when everything is working correctly, this can introduce a short pause between the user’s last input and the assistant’s reply. There isn’t anything misconfigured on your end, and this behavior is normal under certain conditions.

Thank You for explaining this 👍

How is call lag now @Palak Jain it shoulkd be fine now.