Last updated 06-27-2026
Category:
Reviews:
Join thousands of AI enthusiasts in the World of AI!
Vapi
Vapi is a voice agent platform for building, testing, and deploying phone agents that handle real customer calls at scale. Teams use it for customer support, lead qualification, appointment scheduling, and outbound sales, with sub-500ms latency and orchestration built in. The platform has handled over one billion calls and is used by enterprises including Amazon Ring, Intuit, and ServiceTitan.
The product is developer-first but accessible through a dashboard UI as well. You configure voice, conversation flow, telephony, and integrations in one place, then connect 200+ built-in models or bring your own for speech-to-text, LLM, and text-to-speech. Live tool calling and MCP support let agents fetch data and trigger actions mid-call.
Vapi targets engineering teams shipping production voice agents and enterprises that need compliance, uptime guarantees, and dedicated support. Scale customers get SSO, RBAC, SOC 2, HIPAA, PCI, data residency, and a dedicated account team.
Phone responses land in under 500ms for real-time conversations
200+ built-in models and integrations, or plug in your own API keys
Custom SIP, warm transfers, voicemail detection, and DTMF support
Multi-agent orchestration with squads and intelligent handoffs
Live tool calling and MCP access during active calls
Supports 100+ languages including English, Spanish, and Mandarin
Sub-500ms average latency keeps phone conversations feeling natural.
200+ model integrations plus bring-your-own-key support avoid vendor lock-in.
Trusted by Amazon Ring, Intuit, ServiceTitan, and New York Life at enterprise scale.
Full REST API and dashboard UI serve both developers and business teams.
Compliance add-ons cover HIPAA, zero data retention, SOC 2, and PCI on Scale.
Model provider costs for STT, LLM, and TTS are billed separately on the Build tier.
HIPAA compliance requires a $2,000/month add-on on top of usage fees.
SSO, RBAC, and SOC 2 are only available on the Scale annual contract.
Build tier retains call history for 14 days unless you upgrade.
Can I try Vapi for free?
Yes. Vapi's Build plan is usage-based and includes 60+ call minutes to get started. You can sign up at dashboard.vapi.ai without an annual contract. Model provider costs for STT, LLM, and TTS are passed through separately unless you bring your own API keys.
How much does Vapi charge per call minute?
On Vapi's Build plan, hosting costs $0.05 per call minute and SMS or chat messages cost $0.005 each. Model provider costs for speech-to-text, LLM, and text-to-speech are billed at cost, or $0 if you supply your own API keys.
What AI models does Vapi support?
Vapi offers 200+ built-in model and integration options for transcription, language models, and text-to-speech. You can also connect your own models or self-hosted endpoints via the REST API and server SDKs.
Is Vapi HIPAA compliant?
Vapi supports HIPAA compliance as a paid add-on at $2,000 per month on both Build and Scale plans. The Scale enterprise tier also includes SOC 2, PCI, SSO, RBAC, and data residency options.
What languages does Vapi support for voice agents?
Vapi supports English, Spanish, Mandarin, and 100+ other languages for multilingual voice agents. The homepage demo includes English, Spanish, Italian, French, and multi-lingual modes.
Does Vapi have a REST API?
Yes. Vapi is API-first, with REST endpoints, client SDKs, and server SDKs including a TypeScript package on GitHub. Every platform capability is accessible programmatically, and a dashboard UI is available for non-developers.