Retell AI vs Vapi: pricing, features and which to choose

Plans, prices, deployment options, strengths and limits for Retell AI vs Vapi, taken from each company's pricing page, docs and changelog. Updated October 2026.

Retell AIVapi
Retell AI logo

Retell AI

Voice AI Tools

Retell AI is a voice-agent platform for phone and web conversations, with agent building, testing, APIs, webhooks, analytics, and enterprise controls. Pay-as-you-go voice pricing starts at $0.07 per minute before telephony and add-ons, with no platform fee, while enterprise pricing is custom. Its pricing and usage limits depend on the selected subscription and production workflow.

Vapi logo

Vapi

Voice AI Tools

Vapi is an orchestration platform and API for building AI voice agents that answer and place phone calls. It suits developers and enterprises who want to pick their own speech, language and voice models. Usage starts at $0.05 a minute for hosting plus provider costs, with $5 in free credit.

Quick comparison

FeatureRetell AIVapi
CategoryVoice AI ToolsVoice AI Tools
Main useAI phone agentsVoice agents
Pricing modelPay per usePay per use
Free plan Yes Yes
Entry paid plan$0.07-$0.31 per voice minute (Pay-as-you-go)$999/month minimum (Pro)
PlatformsWeb and APIWeb & API
Open source No No
Self-hosted No No
Public API Yes Yes
Profile updatedOctober 2026October 2026

Pricing plans

Every public plan, as listed on each company's pricing page.

Retell AI

  • Pay-as-you-go$0.07-$0.31 per voice minute

    No platform fee or minimum; 20 concurrent calls included; telephony and add-ons may cost extra.

  • EnterpriseCustom pricing

    Higher concurrency, custom terms, SSO, role-based access, dedicated support, and optional dedicated infrastructure.

Vapi

  • Pay as you go$0.05/minute

    Vapi hosting fee plus model costs at no markup. $5 free credit, 4 concurrent calls, 1 phone number, 14-day data retention, community support.

  • CorePrepaid package

    Prepaid, auto-renewing credits. 10 concurrent calls, 5 phone numbers, 30-day retention, zero data retention, email support with a 2 business day SLA.

  • Pro$999/month minimum

    Priced at 10% of hosting fees. 10 organizations, 30 concurrent calls, 180-day retention, RBAC, dedicated Slack support and a 99% uptime SLA.

  • PremierCustom

    Custom concurrency and phone numbers, SSO, dedicated SIP, named support team, one-hour P0 response and 99.9% uptime SLA.

Strengths and limits

Retell AI

  • Retell provides agent building, APIs, web calls, phone calls, testing, analytics, and integrations.
  • The pay-as-you-go tier has no platform fee, contract, or minimum commitment.
  • The platform supports versioning, simulations, transfers, guardrails, multilingual agents, and quality analysis.
  • Retell publishes enterprise options covering security, compliance, retention, and access controls.
  • Final voice cost depends on infrastructure, model, voice, telephony, testing, and add-ons.
  • Testing uses applicable production rates rather than a separate free sandbox.
  • Enterprise pricing, capacity, and some compliance terms require a sales discussion.

Vapi

  • Bring any transcriber, LLM or voice, including custom model endpoints, without rebuilding the agent.
  • Provider costs pass through at cost, so you see what each component charges.
  • Proven at scale, with more than 1 billion calls handled and Amazon Ring in production.
  • Built-in call simulation and monitoring catch failures before customers hear them.
  • Per-minute cost varies with model choice and is hard to forecast early on.
  • The Pro package's $999 monthly minimum is a big jump from pay as you go.
  • Configuration and debugging assume developer skills and comfort reading call logs.

Which should you choose?

Choose Retell AI if

  • Teams building production phone agents with APIs and webhooks.
  • Contact centers needing testing, analytics, transfers, and enterprise controls.
  • Developers comparing language models, voices, telephony, and transcription choices.
  • Organizations reviewing retention, PII, BAA, DPA, or SOC 2 requirements.

Look elsewhere if

  • Buyers seeking one flat voice-generation price.
  • Teams requiring a fully self-hosted voice-agent stack.
  • Projects unable to budget for paid testing and metered components.

Choose Vapi if

  • Engineering teams building phone agents that need to swap speech, language and voice providers.
  • High-volume support lines that already handle thousands of inbound calls a day.
  • Agencies building custom voice receptionists and sales callers for many clients.
  • Healthcare and finance teams that need HIPAA, PCI and zero data retention options.

Look elsewhere if

  • Small businesses without a developer who want a receptionist running the same day.
  • Buyers who need one fixed per-minute price for budgeting.
  • Teams that want a single vendor to own the speech models end to end.

Retell AI logo Retell AI in brief

Retell AI began as a five-person team accepted into Y Combinator in November 2023, then launched publicly in February 2024 through Hacker News and Product Hunt. The company raised a $4.6 million seed round in May 2024, backed by Alt Capital and angel investors; TechCrunch reported $4.53 million raised and hundreds of paying customers that month. Retell’s platform lets businesses build voice agents for phone and web conversations, with templates, APIs, transcripts, analytics, webhooks, simulation tools, and transfers. Its pay-as-you-go plan has no platform fee or minimum commitment, lists voice pricing at $0.07 to $0.31 per minute, and includes 20 concurrent calls. Retell later reported $40 million in annualized revenue in 2025 with a 25-person team, followed by $80 million by December 2025.

A typical Retell workflow begins with a prebuilt template or new agent, followed by configuration of prompts, conversation flows, knowledge sources, tools, transfers, voice settings, and language models. Teams can connect a web call or phone number, then review behavior in Playground, simulation, batch, web, or phone tests. Text tests are billed per message, while web and phone tests use applicable production rates. Before launch, developers can test webhook and custom-function payloads, configure fallback behavior, and revise agent versions. In production, call transcripts, analytics, filters, alerts, and quality tools support monitoring. Agents can also hand calls to staff, use knowledge bases, send SMS, and apply safety guardrails. Retell’s platform supports GPT, Claude, Gemini, ElevenLabs, Cartesia, and other voice or model options.

Retell AI’s main limitation is cost complexity: its $0.07 to $0.31 headline range excludes some telephony and add-on charges, and testing is billed at production rates rather than through a free sandbox. Vapi is a strong alternative for developer-led voice orchestration, while Bland AI offers a more focused phone-agent product; ElevenLabs Conversational AI may appeal more to teams prioritizing voice quality, since TechCrunch’s 2024 test found Retell’s synthetic voice less realistic than ElevenLabs and OpenAI’s. Retell counters with broader workflow tooling, including simulations, webhooks, analytics, transfers, versioning, guardrails, and 20 included concurrent calls. Its enterprise plan adds custom terms, SSO, access controls, and dedicated support, but requires a sales process. Pick Retell for configurable production workflows and metered experimentation, not flat pricing or self-hosting.

Read the full Retell AI review

Vapi logo Vapi in brief

Vapi is a San Francisco company that sells the plumbing behind AI phone agents. Jordan Dearsley and Nikhil Gupta, classmates from the University of Waterloo, went through Y Combinator with a productivity app called Superpowered. In 2023 Dearsley built an AI therapist to talk to on walks, found that startups wanted the low-latency voice layer underneath it, and the pair relaunched as Vapi in 2024. In May 2026 it raised a $50 million Series B led by Peak XV at about a $500 million valuation, bringing total funding to $72 million, according to TechCrunch. The company says it has handled more than 1 billion calls and has over 1 million developers on its self-serve platform. Enterprise customers include Amazon Ring, which routes all of its inbound support calls through Vapi, plus Intuit, New York Life and Kavak.

A Vapi agent is a configuration that wires together three parts: a transcriber that turns caller speech into text, a language model that decides what to say, and a voice that speaks the reply. You set these up in the dashboard or through the API, write a system prompt, and attach tools so the agent can book appointments, look up orders or transfer to a human. Vapi supplies phone numbers, or you bring your own through Twilio, Vonage or SIP. Squads let several specialised agents hand a call to each other. Testing and simulation features run scripted calls against an agent before it reaches customers, and observability dashboards flag failed calls. Web and mobile SDKs cover in-app voice. Common uses are customer support lines, lead qualification, appointment scheduling and outbound sales calls.

Vapi expects you to write prompts, configure tools and read logs, so non-technical teams usually need a developer or an agency to get an agent into production. Total cost is harder to predict than a flat per-minute rate because it depends on which models you pick. Retell AI offers a similar developer platform with simpler bundled pricing and strong turn-taking, and Bland AI targets enterprises that want a more managed service with its own hosted models. ElevenLabs Agents is the easier choice if you already use ElevenLabs voices and want one vendor. Pick Vapi if you have engineers, expect high call volumes and want freedom to change speech, language and voice providers as models improve. Choose Retell or a no-code tool if your team wants a working receptionist this week without touching an API.

Read the full Vapi review

Frequently asked questions

How do Retell AI vs Vapi compare on price?

Retell AI has a free option, and its first paid plan (Pay-as-you-go) is listed at $0.07-$0.31 per voice minute. Vapi has a free option, and its first paid plan (Pro) is listed at $999/month minimum. Prices are from each company's pricing page and can change, so check before you buy.

Which has a free plan, Retell AI vs Vapi?

All of them. Retell AI: No platform fee or minimum; 20 concurrent calls included; telephony and add-ons may cost extra. Vapi: Vapi hosting fee plus model costs at no markup. $5 free credit, 4 concurrent calls, 1 phone number, 14-day data retention, community support.

Can I self-host or integrate Retell AI vs Vapi?

None of them is open source. All of them run as hosted services only. Retell AI and Vapi offer a public API.

What does Retell AI cost?

The pay-as-you-go plan has no platform fee and lists voice pricing at $0.07 to $0.31 per minute. Infrastructure, language model, text-to-speech, telephony, and optional features contribute separate charges. Twenty concurrent calls are included. Phone numbers, extra concurrency, testing, knowledge bases, and other add-ons can add charges. Enterprise pricing is custom.

Is Retell AI free?

Retell allows pay-as-you-go signup without a platform subscription, contract, or minimum commitment. Usage is not free: production calls and testing are metered. Text tests bill per message, while web and phone tests bill per minute at applicable production rates. The included allowance covers 20 concurrent calls, not unlimited usage.

How much does Vapi cost per minute?

Vapi charges a $0.05 per minute hosting fee plus the cost of the transcriber, language model and voice you choose, which it passes through without markup. With Deepgram, an OpenAI model and an ElevenLabs voice, its pricing page puts a typical call at about $0.09 to $0.13 a minute, before phone carrier fees. Pro packages add a $999 monthly minimum.

Does Vapi have a free plan?

Vapi has no permanent free plan, but new accounts get $5 in credit with no success package required. That covers a few dozen test minutes depending on model choice. The starter setup includes one Vapi phone number, four concurrent calls, 14 days of data retention, the full model catalog and community support through Discord. After the credit runs out you prepay for minutes.

Related comparisons

Sources