High-Volume Call Capacity & Multi-Agent Token Opti...
# support
v
Hello VAP Support team, I have two questions about using your VAP system in a high-volume hospital environment: Capacity & Infrastructure We expect around 800 calls per day (≈40 calls/hour), each lasting about 2.5 minutes, for a total of 40 000–45 000 minutes per month (based on our previous human-agent setup). Can VAP reliably handle this call volume? What server or cloud infrastructure do you recommend (e.g. VPS size, cloud provider, hosting location) to ensure smooth performance at this scale? Token Efficiency via Multiple Agents In our text-based bots, we split prompts across several agents (e.g. “Sales,” “Finance,” “Appointments”) and transfer conversations between them to reduce token usage. Is it possible to configure VAP voice agents in the same way—using multiple smaller prompts/agents and transferring a call from one agent to another—to optimize token costs? Thank you for your help! Best regards, Victor Cardoso
k
Yes, Vapi can reliably handle 800 daily calls with sub-500 ms latency and enterprise-grade scale, for smooth performance, use a cloud setup like, AWS/GCP with 2–4 vCPUs, 8 GB RAM, and autoscaling in a nearby region. You can also optimize token usage by using Vapi’s Squads feature to split tasks across specialized agents and transfer calls contextually between them..
v
Thank you, my friend, and one more question: do you know where I can find more advanced material to learn about the possible capabilities of VAPI?
k
But I am exploring the VAPI blog
a
Here is the fixed link -->
k
@Victor Cardoso
2 Views