Skip to content
AITrendTool

Vapi

Developer platform to build, test, and deploy AI voice agents for phone calls with usage-based per-minute pricing

Vapi is an API-first platform for building AI voice agents that handle phone calls. Its orchestration fee is $0.05/minute with no upfront platform cost on the Build plan, and new users get a one-time credit to test. You bring your own speech-to-text, LLM, and text-to-speech providers, whose costs stack on top — so real per-minute spend often lands around $0.07–$0.25+. Best for developers shipping production voice automation.

Verified JUN 24, 2026 USAGE-BASED Live
Screenshot of Vapi

What is Vapi?

Vapi is a developer platform for building AI voice agents that make and take phone calls. It handles the hard parts of real-time voice — telephony, low-latency streaming, turn-taking, and orchestration between speech-to-text, a language model, and text-to-speech — and exposes them through an API you configure to define the agent’s voice, behavior, and conversation flow. The platform targets sub-500ms latency so conversations feel natural rather than stilted.

Pricing is usage-based: Vapi’s own orchestration fee is $0.05 per minute on the self-serve Build plan, with no upfront platform cost and a one-time credit for new accounts. Crucially, you bring your own provider keys for the speech and language components, which avoids markup but means those costs stack on top — real-world spend commonly lands around $0.07 to $0.25 or more per minute. For regulated and high-volume deployments, the enterprise Scale tier adds SOC 2, HIPAA, PCI, SSO, and RBAC under a custom annual contract.

Who is it for?

Vapi is built for developers and teams that want to ship production voice automation rather than experiment with a no-code toy. It rewards users comfortable assembling and managing a provider stack, and it scales from a single agent to enterprise call volumes.

  • Developers and startups building voice products — receptionists, support bots, or outbound callers — who want infrastructure handled but full control over the model stack.
  • Agencies delivering voice agents to local businesses on a build-plus-retainer model.
  • Sales and support teams automating high-volume inbound or outbound calling that would otherwise require staffing.
  • Regulated organizations in healthcare or finance that need compliance controls available on the enterprise Scale tier.

How much does Vapi cost?

Starting price: $0.05/min · Free tier: no · Model: usage-based

Pricing verified JUN 24, 2026

Price history tracked from June 2026

Vapi pricing tiers, verified against the official pricing page
Plan Price Includes
Build $0.05/min No platform fee — pay per minute of calls · 10 concurrent calls included ($10 per extra line/month) · Bring-your-own provider keys (STT/LLM/TTS) · Community Discord and email support
Scale Custom Fixed platform fee plus committed volume · Volume-based per-minute rates · SOC 2, HIPAA, PCI, SSO, and RBAC · Dedicated account team — contact sales

What are Vapi's key features?

  • API-first platform to build, test, and deploy configurable voice agents
  • Managed telephony and infrastructure with a sub-500ms latency target
  • Bring-your-own provider keys for speech-to-text, LLM, and text-to-speech
  • AI guardrails to reduce hallucinations
  • Real-time monitoring plus call and chat history logs
  • Enterprise security: SOC 2, HIPAA, PCI, SSO, OAuth, and RBAC

What people use Vapi for

  1. 01 Automating inbound customer support and call routing
  2. 02 Outbound lead qualification and sales follow-up calls
  3. 03 Appointment scheduling and reminder calls
  4. 04 Handling high-volume phone lines that would otherwise need human staff

Pros and cons

Pros and cons of Vapi
Pros Cons
Transparent low base orchestration fee ($0.05/min) with no upfront platform cost on Build Provider costs (STT + LLM + TTS) and telephony stack on top of the base fee, so effective cost often runs $0.07–$0.25+/minute
Bring-your-own provider keys avoid markup on speech and LLM costs No ongoing free tier — only a one-time trial credit
Strong enterprise and compliance posture (SOC 2, HIPAA, PCI, SSO/RBAC) on Scale Compliance add-ons are expensive and Scale pricing is opaque (annual contract, contact sales)

What are the best Vapi alternatives?

See all Vapi alternatives →

How people make money with Vapi

  • Launch a voice-AI agency building inbound receptionists and appointment-booking agents for clinics, salons, and home-services businesses, charging a build fee plus a monthly management retainer per deployed agent
  • Offer outbound lead-qualification and appointment-reminder voice agents to sales teams on a per-seat or per-location subscription, marking up the underlying per-minute usage you operate on their behalf

Frequently asked questions

How much does Vapi cost?

Vapi charges a $0.05/minute orchestration fee on the Build plan with no upfront platform cost. New users get a one-time trial credit. Speech-to-text, LLM, and text-to-speech provider costs are billed separately on top of that base fee.

Does Vapi have a free tier?

Vapi does not offer an ongoing free tier. New accounts receive a one-time credit to test the platform, after which you pay the $0.05/minute base fee plus your chosen provider costs. The enterprise Scale tier is contact-sales only.

What is the real per-minute cost of a Vapi agent?

The $0.05/minute is only Vapi's orchestration fee. Once you add speech-to-text, an LLM, text-to-speech, and telephony, the effective cost commonly lands around $0.07 to $0.25 or more per minute, depending on which providers you choose.

Can Vapi handle multiple calls at once?

Yes. The Build plan includes 10 concurrent calls, with additional lines available at $10 per line per month. The enterprise Scale tier supports higher committed concurrency under a custom annual contract.

Is Vapi suitable for healthcare or regulated industries?

Vapi offers compliance options including SOC 2, HIPAA, PCI, SSO, and RBAC on its Scale tier. These features sit behind enterprise contracts and paid add-ons, so regulated deployments require the Scale plan rather than the self-serve Build plan.

Do I need to bring my own AI models to Vapi?

Vapi is designed around bring-your-own provider keys for speech-to-text, the LLM, and text-to-speech. This avoids markup but means you manage and pay those providers directly, and your total cost depends on the stack you assemble.