Vapi pricing: hosting, providers and the complete monthly bill
Understand the difference between Vapi hosting charges and provider, carrier, package and operating expenses.
Table of Contents▼
Vapi's published hosting fee is $0.05 per minute, with model providers accounted for separately. That hosting figure is therefore a starting point, not a complete cost for a telephone conversation. Check the selected configuration on the current pricing page.
Estimate one configuration at a time
Write down the transcriber, language model, voice, telephone route and account package you intend to use. If you switch to a realtime speech model, revisit the billing units rather than reusing a cascade estimate unchanged.
Distinguish charges collected by the platform from charges paid directly to your provider accounts. Both belong in the total, but neither should be counted twice.
A simple hosting example
At the published hosting rate above, 3,000 billable minutes would create $150 in hosting charges before other applicable costs. This is arithmetic for that one component, not an all-in Vapi quote.
To complete the forecast, add actual provider usage estimates, carrier charges, rented numbers and selected package or add-on expenses. For model components billed in tokens or characters, minute-based estimates depend on how much the caller and assistant speak and how much context the model receives.
Questions that change the bill
| Question | What to establish |
|---|---|
| Does the call include silence or hold time? | Which connected intervals remain billable |
| Will calls transfer? | Duration and cost of every telephone leg |
| Are provider keys supplied directly? | Where those charges appear |
| Is additional capacity needed? | Applicable package or concurrency expense |
| How long is data retained? | Included retention and any selected changes |
A plan's included capacity is not the same as the traffic your application can safely process. Test dependencies such as calendars and CRMs as well as the voice system.
Reconcile before optimizing
Pick a completed call and connect its identifier to the platform usage, carrier duration and provider charges. Compare a normal call with one that generated an unexpectedly large expense. Long responses, repeated work or long connected periods are hypotheses to inspect, not proof of an incorrect invoice.
Fix avoidable work first, then compare provider settings with the same scripts. Track cost per useful result so a cheaper configuration does not quietly increase failures.
Comparing Vapi with Burki
Use the same workload and include all separately charged components on both sides. Burki's current pricing should not be compared with a Vapi total as if a platform-only rate covers the same services. The complete pricing worksheet provides a consistent structure for that decision.
Ready to try Burki?
Create an assistant and check your available browser practice allowance.
Create your assistantTrial eligibility and available practice are shown in your workspace.