What exactly is included in the per-minute cost?

The per-minute cost is the sum of your chosen Transcriber (e.g., Deepgram), Intelligence Model (e.g., OpenAI), and Voice (e.g., ElevenLabs) provider rates. Vapi charges no markup on these.

How is the Pro package price calculated?

The Pro package costs 10% of your total Vapi hosting fee, with a $999/month minimum. The hosting fee is based on your per-minute usage multiplied by the base rate.

Can I use any STT/LLM/TTS provider with Vapi?

Yes. Vapi offers a full catalog of providers like OpenAI, Anthropic, Deepgram, and ElevenLabs. You can also bring your own provider keys for custom or self-hosted models.

How do Squads differ from regular Assistants?

Assistants are single-prompt agents. Squads orchestrate multiple assistants, allowing context-preserving handoffs between specialized agents (e.g., triage to scheduling).

What compliance certifications does Vapi have?

Vapi is SOC 2 compliant. HIPAA and PCI compliance are available as enterprise add-ons. They provide a Data Processing Agreement and Business Associate Agreement.

What is the typical latency for a voice agent response?

Vapi targets sub-500ms average latency for enterprise clients. Performance depends on your chosen providers; the platform optimizes the orchestration pipeline.

Directory/Chatbots/AI Assistants/Vapi

Vapi

Vapi is the developer platform for building voice AI agents. It handles complex infrastructure so you can focus on creating great voice experiences, allowing you to build agents that have natural conversations, make and receive phone calls, and integrate with existing systems. Key capabilities include real-time conversations with sub-600ms response times, phone and web integration, tool integration to connect to your APIs and databases, and multi-assistant orchestration with Squads for specialized workflows. The platform combines three core technologies: speech-to-text, large language models, and text-to-speech, giving you full control over each component with dozens of providers to choose from. You can build voice agents using either Assistants for most use cases or Squads for multi-assistant setups. Vapi is used by enterprises to automate customer support, qualify leads, handle scheduling, and manage complex workflows like medical triage or e-commerce operations.

Chatbots

Visit website

Added on

Sep 22, 2026

Monthly Visits

951K

Voice AI agentsPhone & web callsMulti-assistant SquadsReal-time monitoringAPI-nativePay-per-minuteHIPAA/SOC2

Product information

Everything worth knowing about Vapi

The complete picture — from everyday features to the technical detail builders and IT folks dig for.

Core features

What it actually does

Voice Agent Orchestration

Configure and deploy voice AI agents for phone and web conversations, handling infrastructure from telephony to real-time response.

Multi-Assistant Orchestration (Squads)

Compose specialized assistants for complex workflows like medical triage or e-commerce, with context-preserving handoffs between agents.

Unified Monitoring & Analytics

Track call performance, surface issues, and continuously improve agent experiences through real-time observability and structured outputs.

Extensive Integration Ecosystem

Connect to leading providers (OpenAI, Anthropic, ElevenLabs, Deepgram) and third-party tools (Zendesk, HubSpot, Zapier) via API-native design.

Pay-as-You-Go Usage Model

Pay only for minutes consumed, with transparent per-component costs (transcription, intelligence, voice, telephony) and optional success packages.

Technical capabilities

Telephony & SIP Integration

Provides Vapi-managed phone numbers, inbound/outbound telephony via Twilio/Vonage/Telnyx, and dedicated SIP for enterprise deployments.

Model Intelligence Presets

Pre-configured bundles of transcriber, LLM, and voice models optimized for common use cases, enabling one-click assistant setup.

Must watch videos

Build a Voice Agent in 15 Minutes Using VAPI (Easiest Tutorial)

15:42

Tutorial15:42

Build a Voice Agent in 15 Minutes Using VAPI (Easiest Tutorial)

https://vapi.ai/?aff=intellagents-ZVSDm5siKr4 (*VAPI $10 FREE) In this VAPI tutorial, I'll walk you through creating an AI receptionist from scratch in just 15 minutes. You'll learn the basics of setting up your first AI agent, including initial conversation flows and integrating with essential tools like Google Calendar to manage bookings and appointments automatically. Beyond the initial setup, we'll delve into optimising your voice agent for real-world scenarios. Discover how to handle edge cases gracefully, such as transferring calls to a human agent or suggesting alternative times when a slot is unavailable. We'll also cover crucial aspects like crafting effective global prompts, using variables for better data capture, and selecting the right AI models (like GPT-4.1) to balance cost, quality, and latency. By the end, you'll understand why VAPI stands out among other voice AI tools for its affordability and impressive performance, making it one of the easiest ways to build voice AI. *🔗 Links & Resources Mentioned* Compare Voice AI Tools: https://voiceai.guide VAPI Pricing Page: https://vapi.ai/?aff=intellagents-ZVSDm5siKr4 If you run into any issues or have a question drop me a comment and I'll do my best to help you get up and running. *Why watch?* If it's your first view — hey 👋 I'm Marcus! I spent 12 years building an 8-figure software venture studio (we built 12 businesses – 6 failed, 4 held, 2 exited). I get a kick out of automating businesses and I'm now sharing my journey learning about AI automation in real time to help and connect with others in the space. My channel's still new and small, so your like/comment/sub genuinely helps! 👉 Subscribe: https://www.youtube.com/@ventureharbourtestlab?sub_confirmation=1 *⏱️ Timestamps* 0:00 - Video Overview 0:14 - Core Workflow 3:15 - Configure Calendar 5:22 - First Agent Test 6:30 - Handle Edge Cases 8:22 - Test Edge Cases 9:30 - Advanced Vapy Tips 12:35 - Pricing Analysis 14:21 - Performance & Integrations *🍿 WATCH NEXT* ▶️ I Built the Same Voice Agent on VAPI, Synthflow & Retell to find the best: https://www.youtube.com/watch?v=up2oklqQuUE ▶️ How to Build an AI Receptionist in 10 Minutes: https://www.youtube.com/watch?v=WKA8fIW76nY ▶️ How I Automated Phone Calls With AI (Step-by-Step, No-Code): https://www.youtube.com/watch?v=4SL89yfFGLg #VAPI #VoiceAI #AIVoiceAgent

CloudTalk vs Vapi: Which AI Voice Agent Actually Handles Calls Better in 2026?

11:48

Comparison11:48

CloudTalk vs Vapi: Which AI Voice Agent Actually Handles Calls Better in 2026?

🎙️ GET CLOUDTALK: 15% off your first three months: https://go.davidalex.com/cloudtalk 🛠️ GET VAPI: https://go.davidalex.com/vapi If you're adding AI voice agents to your business and you're not a developer, most tools in this space weren't built for you. I broke down CloudTalk and Vapi side by side — the core differences, how the features stack up, and what the pricing actually looks like once you factor in everything you'll need. These two aren't even in the same category of product: CloudTalk is an all-in-one business phone system with AI baked in, while Vapi is a developer-first orchestration layer you assemble yourself. Walking through CloudTalk's phone system + AI agent (160+ countries, 60+ languages, HubSpot and Salesforce integrations built in) against Vapi's bring-your-own-everything approach, including the pricing trap most people miss — Vapi's "5 cents a minute" only covers the API, and stacking on your LLM, voice, transcription, and telephony provider pushes real cost to 15-30 cents a minute. By the end, you'll know exactly which one fits your use case. 🎙️ GET CLOUDTALK: 15% off your first three months: https://go.davidalex.com/cloudtalk 🛠️ GET VAPI: https://go.davidalex.com/vapi ⏱️ TIMESTAMPS: 0:00 - Not a Developer? 1:06 - What CloudTalk Is 2:50 - What Vapi Is 4:01 - One Platform vs a Stack 6:04 - CRM Integrations 8:06 - The Real Vapi Cost 9:36 - CloudTalk Pricing 10:50 - Which One Wins

I Built an AI Voice Receptionist with Vapi and n8n MCP (free template)

34:21

Tutorial34:21

I Built an AI Voice Receptionist with Vapi and n8n MCP (free template)

My FREE AI OS Course: https://www.skool.com/ai-automation-society/about?el=i-built-an-ai-voice-receptionist-with-vapi-and&hcategory=youtube-videos&utm_campaign=free-group Full courses + unlimited support: https://www.skool.com/ai-automation-society-plus/about?el=i-built-an-ai-voice-receptionist-with-vapi-and&hcategory=youtube-videos&utm_campaign=ais-plus Apply for my YT podcast: https://podcast.nateherk.com/apply Work with me: https://uppitai.com/ My Tools💻 FREE MONTH voice to text: https://get.glaido.com/nate Code NATEHERK for 10% off VPS (annual plan): https://www.hostinger.com/vps/claude-code-hosting I built a full AI voice receptionist using VAPI on the front-end and n8n on the back-end, powered by an n8n MCP server that connects to seven n8n workflows. In this video, I walk through the entire system from scratch and show how the agent gathers info, manages the CRM, books and updates appointments, transfers calls, answers questions, logs every call, and keeps an appointment history. Everything is built with no code and designed to be easy to customize for your own use case. By the end, you’ll understand the full setup and have everything you need to launch your own AI receptionist. Sponsorship Inquiries: 📧 nate@smoothmedia.co Connect with me: https://www.linkedin.com/in/nateherkelman/ https://x.com/nateherk https://www.instagram.com/nateherk/ TIMESTAMPS 00:00 Live Demo 06:00 Voice AI Transparency 06:32 High Level Breakdown & Wireframe 10:33 Vapi Model Configuration 11:49 n8n MCP Tool 13:30 Client Lookup and New Client 15:30 Setting up Vapi to ‘Call’ n8n 18:00 Check Availability and Book Appointment 21:30 Update Appointment and Delete Appointment 23:30 Handoff and Transfer Call 25:33 Knowledge Base Files 27:22 Why n8n MCP? 28:50 Logging Call Summaries 32:13 Setting up Phone Numbers 32:57 Get the FREE Resources 33:35 Want to Master AI Automations?

Who uses it

Real use cases, no hype

Customer Support Automation

Deploy inbound voice agents that handle FAQs, ticketing, and escalation to humans.

Sales & Lead Qualification

Automate outbound calls to score leads, book demos, and nurture pipelines.

Appointment Scheduling

Build agents that check calendars, book slots, and send confirmations.

Medical Triage & Scheduling

Use specialized Squads to route emergencies and book patient appointments.

E-commerce Order Management

Handle order tracking, returns, and support queries via voice.

Web Voice Integration

Embed live voice chat into your web app for real-time user interaction.

Enterprise IVR Replacement

Replace legacy phone systems with intelligent, API-native voice agents.

What's great

Rapid deployment from zero to production in weeks
Massive provider ecosystem (OpenAI, Anthropic, ElevenLabs, Deepgram)
Unified platform for phone & web voice agents
Enterprise-grade features: SOC 2, HIPAA, 99.9% SLA
Flexible pricing model: pay-per-minute with no-markup pass-through costs

Technical strengths

Advanced orchestration with 'Squads' for multi-assistant workflows
Sub-500ms latency with dedicated infrastructure scaling
Full control over STT, LLM, TTS pipeline per assistant

Where it falls short

Complex, usage-based pricing can be unpredictable at scale
Steep learning curve for advanced features like Squads
Minimum $999/mo commitment for production-tier Pro package
HIPAA compliance is a costly $2000/mo add-on
Vendor lock-in risk on proprietary platform features

Technical limitations

Real-time monitoring requires separate observability integration
Telephony transport fees from third-party providers add hidden costs

For technical folks

The deep-dive specs

FAQ

Includes technical Q&A

Usage fees are the pass-through costs for minutes consumed (STT, LLM, TTS). A success package is a monthly subscription for platform features, support, SLAs, and data retention.