Voice agents · · 14 min

AI voice agent pricing in 2026: cost per minute vs custom build

About $0.08 to $0.15 a minute on a hosted platform, $0.04 on a custom stack: the line items behind each rate, plan minimums, and the volume where custom pays off.

Quick answer

On a hosted platform, AI voice agent pricing comes to about $0.08 to $0.15 per minute all-in at list prices, plus extras such as HIPAA add-ons. A custom stack’s usage costs about $0.04 per minute, but it needs a $25k–$120k build and its own hosting. Custom starts paying back between 11,000 and 31,000 minutes a month.

Listen to the summary ·

Read the transcript

Vapi advertises $0.05 a minute, but that rate covers only one layer of the call. Add telephony, speech-to-text, the language model and the voice, and a hosted platform costs about $0.08 to $0.15 a minute at list prices, on Vapi or Retell.

A custom stack built on Twilio, Deepgram and Claude costs about $0.04 a minute in usage. It also needs a fixed-price build and its own hosting, so at low volume the platform is cheaper.

On list prices checked October 18, 2026, custom starts paying back somewhere between 11,000 and 31,000 minutes a month, depending on the platform and model you compare against. A HIPAA add-on moves that point down: Vapi lists one at $2,000 a month.

Per-minute quotes also leave costs out: extra concurrent calls, plan minimums, transfers, testing, and the integration work that sits in the build.

For most first voice agents, a platform is the better choice: under about 10,000 minutes a month, with a standard call flow. The guide below shows every line item and calculation, so you can rerun it with your own call volume.

AI-generated voice summary of this guide.

AI voice agent pricing comes to about $0.08 to $0.15 a minute all-in on a hosted platform such as Vapi or Retell, and about $0.04 a minute in usage on a custom stack before hosting and the build. Vapi advertises $0.05 a minute, but that covers one layer of a five-layer stack; add the speech-to-text, the language model and the voice, and its own calculator shows $0.082 to $0.129.

Every figure below comes from a public price list checked on October 18, 2026 (Vapi and Retell rechecked November 10, 2026, unchanged), and every calculation is shown, so you can rerun it with your own call volume.

How much does an AI voice agent cost per minute?

At list prices, a hosted voice platform costs about $0.08 to $0.15 per minute once every layer is included. A custom stack that pays each vendor directly costs about $0.04 per minute in usage, but you have to build and host it. The platform’s advertised rate, $0.05 on Vapi or $0.055 on Retell, is only the orchestration fee.

All-in cost per minute, list prices (checked Oct 18, 2026)
OptionPer minuteWhat is included
Vapi$0.082–$0.129$0.05 hosting plus speech-to-text, model and voice passed through at cost, on Vapi telephony (no transport fee)
Retell, Claude 4.5 Haiku$0.110$0.055 infrastructure, $0.015 voice, $0.025 model, $0.015 telephony
Retell, recommended tier$0.149Same, with a $0.064 recommended-tier model such as Claude 5 Sonnet
Custom stackabout $0.043Twilio, Deepgram Nova-3, Deepgram Aura-2 voice, Claude Sonnet 5.5; hosting and build not included

All-in cost per minute, list prices (checked Oct 18, 2026)

Vapi
Per minute
$0.082–$0.129
What is included
$0.05 hosting plus speech-to-text, model and voice passed through at cost, on Vapi telephony (no transport fee)
Retell, Claude 4.5 Haiku
Per minute
$0.110
What is included
$0.055 infrastructure, $0.015 voice, $0.025 model, $0.015 telephony
Retell, recommended tier
Per minute
$0.149
What is included
Same, with a $0.064 recommended-tier model such as Claude 5 Sonnet
Custom stack
Per minute
about $0.043
What is included
Twilio, Deepgram Nova-3, Deepgram Aura-2 voice, Claude Sonnet 5.5; hosting and build not included

The five line items inside every per-minute price

Every AI phone call pays for the same five things, whether a platform bundles them or you buy them separately. Knowing them is how you compare quotes.

  • Telephony carries the call. Twilio lists US inbound calls to a local number at $0.0085 a minute, and streaming the audio to your server at $0.0044 a minute.
  • Speech-to-text transcribes the caller. Deepgram lists streaming Nova-3 at a regular price of $0.0077 a minute, currently discounted to $0.0048; we use the regular price.
  • The language model decides what to say and which tools to call. It is billed per token, so its per-minute cost depends on prompt size and how often the caller speaks.
  • Text-to-speech speaks the reply. Deepgram lists Aura-2 at $0.030 per 1,000 characters; Retell lists voices from $0.015 to $0.10 a minute depending on the provider.
  • Orchestration handles turn-taking, interruptions and tool calls. Platforms charge for it per minute; on a custom stack you host it yourself.
Two stacked bars on a scale of 0 to 12 cents a minute. Retell with Claude 4.5 Haiku, $0.110: infrastructure $0.055, model $0.025, voice $0.015, telephony $0.015. Custom stack, about $0.043: Twilio call and stream $0.0129, speech-to-text $0.0077, model $0.0103, voice $0.012, excluding hosting and the build.Two stacked bars on a scale of 0 to 12 cents a minute. Retell with Claude 4.5 Haiku, $0.110: infrastructure $0.055, model $0.025, voice $0.015, telephony $0.015. Custom stack, about $0.043: Twilio call and stream $0.0129, speech-to-text $0.0077, model $0.0103, voice $0.012, excluding hosting and the build.
On Retell the platform fee is half the minute; on a custom stack you host orchestration yourself. Model and voice per-minute figures use the assumptions below.

How we turned token prices into a per-minute cost

Two of these layers are not priced per minute, so we converted them with stated assumptions. For the model, we assume five model calls per minute of conversation, each sending 3,000 input tokens (2,500 of them a cached system prompt) and receiving 80 output tokens. At Claude Sonnet 5.5 list prices of $2 per million input tokens, $0.10 for cache reads (Anthropic bills cache hits on Sonnet 5.5 at 0.05 times the input price) and $10 output, that is about $0.0021 per call and $0.0103 a minute. For the voice, we assume the agent talks for about 45% of the call at around 150 words a minute, roughly 400 characters, which is $0.012 a minute on Aura-2.

AI voice agent platform pricing: what Vapi and Retell charge

Both platforms publish their prices, so a fair comparison is possible. They differ in what is bundled and what costs extra. Other platforms bundle and bill differently: Bland charges per talk minute with the model and voice included, and Google’s Dialogflow CX bills per second of audio. We track Vapi, Retell, Bland and other vendors’ list prices, each with its source and check date, and recheck them monthly.

Platform price lists compared (checked Oct 18, 2026)
VapiRetell
Platform fee$0.05/min hosting$0.055/min voice infrastructure
Models and voicesPassed through at cost; calculator examples $0.0095–$0.0099 transcriber, $0.0077–$0.0452 model, $0.0146–$0.0238 voicePriced per minute: voices $0.015–$0.10; models from under $0.01 to over $0.30, with Claude 4.5 Haiku at $0.025 and the recommended tier at $0.064
TelephonyVapi telephony and SIP free, one number on the free tier; Twilio numbers billed by Twilio$0.015/min, numbers $2 a month
PlansNo Success Package $0 with $5 credit; Core $29 a month; Pro 10% of hosting, $999 a month minimum; Premier customUsage-based, $10 in free credits; enterprise plan at custom pricing
Concurrent calls4 / 10 / 30 by plan, extra lines $10 each a monthFirst 20 free, then $8 each a month
HIPAA$2,000 a month add-onHIPAA and BAA listed on pay-as-you-go and enterprise; custom BAA on enterprise (checked Oct 19, 2026)

Platform price lists compared (checked Oct 18, 2026)

Platform fee
Vapi
$0.05/min hosting
Retell
$0.055/min voice infrastructure
Models and voices
Vapi
Passed through at cost; calculator examples $0.0095–$0.0099 transcriber, $0.0077–$0.0452 model, $0.0146–$0.0238 voice
Retell
Priced per minute: voices $0.015–$0.10; models from under $0.01 to over $0.30, with Claude 4.5 Haiku at $0.025 and the recommended tier at $0.064
Telephony
Vapi
Vapi telephony and SIP free, one number on the free tier; Twilio numbers billed by Twilio
Retell
$0.015/min, numbers $2 a month
Plans
Vapi
No Success Package $0 with $5 credit; Core $29 a month; Pro 10% of hosting, $999 a month minimum; Premier custom
Retell
Usage-based, $10 in free credits; enterprise plan at custom pricing
Concurrent calls
Vapi
4 / 10 / 30 by plan, extra lines $10 each a month
Retell
First 20 free, then $8 each a month
HIPAA
Vapi
$2,000 a month add-on
Retell
HIPAA and BAA listed on pay-as-you-go and enterprise; custom BAA on enterprise (checked Oct 19, 2026)

The choice of model and voice moves the platform price more than the choice of platform. On Retell, swapping a recommended-tier model for Claude 4.5 Haiku takes the all-in rate from $0.149 to $0.110 a minute, the most expensive models on Retell’s list cost over $0.30 a minute on their own, and moving to a premium ElevenLabs voice can add up to $0.085.

A third route: one speech-to-speech model

Some stacks replace the separate speech-to-text, model and voice with one speech-to-speech model that hears and answers in audio. OpenAI’s realtime models are billed per audio token rather than per minute ($32 per million input and $64 per million output tokens for gpt-realtime-2.1; $10 and $20 for the mini model, list prices checked October 26, 2026). In our own test calls in October 2026, that came to about $0.078 a minute of talk for gpt-realtime-2.1 and $0.023 for the mini model, before telephony, hosting and orchestration; the method and raw data are in our voice agent latency benchmark. Long system prompts and tool definitions add text tokens to every turn, so measure with your own call flow before you budget.

Per minute, per call or subscription: comparing pricing models

Most AI voice agents are billed per minute, but some vendors sell bundles of minutes or charge per conversation, and those rates only compare once you convert them to a cost per minute. The table does that for one vendor of each model, at list prices checked November 10, 2026, and a three-minute average call.

Pricing models converted to cost per minute (checked Nov 10, 2026)
ModelExampleList pricePer minute
Per minute, components at costVapi$0.05 hosting plus transcriber, model and voice$0.082–$0.129 all-in
Per minute, priced per componentRetell$0.055 infrastructure plus voice, model and $0.015 telephony$0.110–$0.149 all-in
Subscription with included minutesElevenAgents Pro$99 a month for 1,238 minutes, extra minutes $0.08$0.08, plus the model, billed separately
Per conversationRingly$99 a month minimum covering 50 conversations, then $1.99 eachabout $0.66 at three minutes a call

Pricing models converted to cost per minute (checked Nov 10, 2026)

Per minute, components at cost
Example
Vapi
List price
$0.05 hosting plus transcriber, model and voice
Per minute
$0.082–$0.129 all-in
Per minute, priced per component
Example
Retell
List price
$0.055 infrastructure plus voice, model and $0.015 telephony
Per minute
$0.110–$0.149 all-in
Subscription with included minutes
Example
ElevenAgents Pro
List price
$99 a month for 1,238 minutes, extra minutes $0.08
Per minute
$0.08, plus the model, billed separately
Per conversation
Example
Ringly
List price
$99 a month minimum covering 50 conversations, then $1.99 each
Per minute
about $0.66 at three minutes a call

Per-minute billing is usually cheapest once volume is steady, and you pay only for the minutes you use. A subscription buys a predictable bill, but unused minutes are lost and ElevenAgents charges $0.16 a minute for burst calls above your plan’s concurrency. Per-conversation pricing suits a small business that wants a finished AI receptionist rather than a platform: Ringly does not bill spam, hang-ups or callers who only want a person, and a long call costs the same as a short one. At $1.99 a conversation it beats a $0.11 all-in rate only when calls average about 18 minutes.

Voice AI development costs: what a custom agent costs to build

A custom voice agent costs $25k–$120k to build with us, fixed after a $5,000 discovery sprint. Published market ranges are wider: RaftLabs (updated June 7, 2026) puts a basic IVR replacement at $15,000 to $35,000, a production agent with several intents and two to four integrations at $35,000 to $80,000, and an enterprise voice platform at $80,000 to $150,000 or more, and calls integration complexity the largest hidden cost.

With us, a custom voice agent falls in the same $25k–$120k band as any agent we build, with the price fixed after discovery (the pricing page has the model). “Custom” means you own the orchestration code, run it in your cloud and pay each vendor directly, typically on an open-source framework such as LiveKit or Pipecat instead of a hosted platform.

Five things set where a voice build lands in that band:

  • the systems the agent books into or updates, such as a scheduler, CRM or order system,
  • the number of call flows, from a single booking flow to triage across several departments,
  • warm transfer to a person with a summary of the call,
  • latency tuning, because slow replies make a conversation feel broken (see voice agent latency),
  • compliance, such as HIPAA, call recording consent and data retention.

A platform build is not free either. Someone still writes the prompts, connects the agent to your systems and tests the call flows. That work is similar on both routes, so the comparison below charges only the custom route with a build, which favors the platforms.

Total cost at 1,000, 10,000 and 50,000 minutes a month

At 1,000 minutes a month a platform costs roughly a tenth of a custom stack; at 50,000 the order flips. The table adds usage at list prices and, for the custom stack, two fixed monthly costs we state as assumptions: $500 a month for hosting and monitoring, and the minimum build of $25k spread over 36 months, about $694 a month.

Monthly cost by call volume (list prices, assumptions above)
Minutes a monthVapiRetellCustom usageCustom all-in
1,000$82–$129$110–$149$43$1,237
10,000$818–$1,289$1,100–$1,490$429$1,623
50,000$4,090–$6,445$5,500–$7,450$2,143$3,337

Monthly cost by call volume (list prices, assumptions above)

1,000
Vapi
$82–$129
Retell
$110–$149
Custom usage
$43
Custom all-in
$1,237
10,000
Vapi
$818–$1,289
Retell
$1,100–$1,490
Custom usage
$429
Custom all-in
$1,623
50,000
Vapi
$4,090–$6,445
Retell
$5,500–$7,450
Custom usage
$2,143
Custom all-in
$3,337

The platform columns exclude plan fees, concurrency charges and HIPAA add-ons, and the custom column excludes the cost of your team’s time to operate it. Both omissions matter at the edges, so treat the table as a starting model, not a quote.

The break-even point for a custom voice agent

On these list prices, a custom stack breaks even at about 11,300 minutes a month against Retell’s recommended tier, 13,900 against Vapi’s higher configuration, 17,800 against Retell with Claude 4.5 Haiku, and 30,700 against Vapi’s lowest configuration. The arithmetic is the custom stack’s fixed monthly cost, about $1,194, divided by the per-minute saving.

Break-even minutes per month for a custom stack
Compared withPlatform per minuteSaving per minuteBreak-even
Retell, recommended tier$0.149$0.106about 11,300
Vapi, higher configuration$0.129$0.086about 13,900
Retell, Claude 4.5 Haiku$0.110$0.067about 17,800
Vapi, lowest configuration$0.082$0.039about 30,700

Break-even minutes per month for a custom stack

Retell, recommended tier
Platform per minute
$0.149
Saving per minute
$0.106
Break-even
about 11,300
Vapi, higher configuration
Platform per minute
$0.129
Saving per minute
$0.086
Break-even
about 13,900
Retell, Claude 4.5 Haiku
Platform per minute
$0.110
Saving per minute
$0.067
Break-even
about 17,800
Vapi, lowest configuration
Platform per minute
$0.082
Saving per minute
$0.039
Break-even
about 30,700

Two things move the break-even sharply, in opposite directions. A HIPAA add-on of $2,000 a month on a platform is larger than the custom stack’s entire fixed cost in this model, so it pulls the break-even down, although a custom stack has its own compliance costs, such as Twilio’s Security Edition for a BAA. A larger build, say for several call flows and integrations, pushes it up by roughly 6,500 to 17,800 minutes for every additional $25k spread over three years, depending on the platform you compare against.

A HIPAA add-on vs the custom stack’s fixed monthly costs
  • Custom: hosting and monitoring$500
  • Custom: minimum build over 36 months$694
  • Vapi HIPAA add-on$2,000

Custom stack assumptions from the total-cost section: about $1,194 a month in all.

What the first year costs in cash

The break-even above spreads the build over three years. Paid up front, the build makes a custom stack more expensive in year one than every platform configuration below about 24,000 minutes a month, and than the cheapest one, Vapi’s lowest configuration, below about 66,000; from year two it pays only hosting and usage. The table uses the same list prices and assumptions as the total-cost section, with the minimum build of $25k paid in year one.

Cash cost per year: platform vs custom (list prices, minimum build)
Minutes a monthVapi, a yearRetell, a yearCustom, year oneCustom, year two on
5,000$4,908–$7,734$6,600–$8,940$33,572$8,572
20,000$19,632–$30,936$26,400–$35,760$41,286$16,286
50,000$49,080–$77,340$66,000–$89,400$56,716$31,716

Cash cost per year: platform vs custom (list prices, minimum build)

5,000
Vapi, a year
$4,908–$7,734
Retell, a year
$6,600–$8,940
Custom, year one
$33,572
Custom, year two on
$8,572
20,000
Vapi, a year
$19,632–$30,936
Retell, a year
$26,400–$35,760
Custom, year one
$41,286
Custom, year two on
$16,286
50,000
Vapi, a year
$49,080–$77,340
Retell, a year
$66,000–$89,400
Custom, year one
$56,716
Custom, year two on
$31,716

At 20,000 minutes a month the custom stack costs more than any platform configuration in year one, and less than every one from year two. At 5,000 minutes it stays more expensive than Vapi even in year two: its $6,000 a year of hosting alone is close to a full year of Vapi usage. So the question for a custom build is not the per-minute rate but whether your volume will still be there in year two.

Costs that per-minute quotes leave out

A per-minute rate is the biggest line on the bill, but not the only one. Budget for these as well, on any route:

  • Concurrent calls. Platforms include a set number of simultaneous calls and charge for more: Retell includes 20 and charges $8 a month for each extra, and Vapi includes 4, 10 or 30 depending on the plan and charges $10 a month per additional line. Size this for your busiest hour, not your average.
  • Plan minimums. Vapi’s Core plan costs $29 a month, and its Pro plan bills 10% of the hosting fee with a $999 a month minimum. At 5,000 minutes a month, Pro’s hosting share is $25, so the minimum, not the percentage, sets the bill until volume is far higher.
  • Phone numbers. Small, but they add up across locations: Twilio lists local numbers at $1.15 a month and toll-free at $2.15.
  • Transfers. When the agent hands a caller to a person, the outbound leg is billed too; Twilio lists outbound calls at $0.0140 a minute.
  • Recordings and transcripts. Storage and retention rules matter more than the storage price, especially for regulated calls.
  • Testing. Every prompt or model change should be replayed against recorded test calls before it goes live, and those test calls use real minutes.
  • Change. Models, voices and prices change; someone has to evaluate and apply those changes.

The integration is the cost that no per-minute quote shows, because it sits in the build. On the service booking agent we built for Innovative AI, every booking call loaded that vehicle’s own menu of 100–200 services from Xtime, checked the dealership’s open slots and settled drop-off, pickup or mobile service before it booked anything. In a build like this, each lookup is typically a tool the agent calls mid-conversation, and how AI service scheduling works at a dealership walks through them in order.

HIPAA and the BAA chain

A voice agent that hears patient information is only as compliant as the weakest vendor on the call. Under HIPAA, every vendor that receives protected health information must sign a business associate agreement, and on a voice call that is the carrier, speech-to-text, the language model, text-to-speech, the orchestration layer and wherever recordings and transcripts are stored.

Price this chain before you choose a stack. Vapi lists HIPAA-eligible data handling at $2,000 a month. Retell’s plan comparison lists HIPAA and a BAA on pay-as-you-go as well as enterprise, with a custom BAA on enterprise only (checked October 19, 2026). Twilio signs a BAA only for customers on its Security or Enterprise Edition. On a custom stack you sign each agreement yourself and can choose vendors and regions accordingly, which is often the deciding factor for healthcare work. There is no such thing as being “HIPAA certified”, so ask each vendor for its BAA, not a badge.

Outbound calls: consent rules change the cost

An agent that places calls costs more to build than one that answers them, because US law treats its voice as artificial. In February 2024 the FCC confirmed, in a declaratory ruling (FCC 24-17), that the TCPA’s restrictions on “artificial or prerecorded voice” cover current AI technologies that generate human voices. Such calls need the called party’s prior express consent, absent an emergency purpose or an exemption.

For a build, that turns into line items an inbound receptionist mostly avoids:

  • consent capture and storage, tied to each number the agent may call,
  • do-not-call list scrubbing before every campaign,
  • an opt-out path inside the call that the agent honors at once,
  • calling-hour limits by the recipient’s time zone,
  • logs that show which consent covered which call.

Budget these as scope, not as a later fix; they decide whether the agent can dial at all. This is a summary of the ruling, not legal advice, and state rules can add to it.

Calls with a sales purpose count as telemarketing and need prior express written consent, so an unsold-lead callback carries a heavier consent flow than a reminder about a booked service, as our guide to no-show and unsold-lead follow-up calls explains.

When a platform is the right answer

For most first voice agents, a platform is the better choice. Choose one when:

  • volume is under about 10,000 minutes a month and not growing fast;
  • the call flow is standard, such as answering questions, booking into a common scheduler or taking messages;
  • you want to test whether callers accept an AI agent before you invest in a build;
  • or nobody will run voice infrastructure after launch.

Go custom when volume is high or rising, when compliance needs control over every vendor in the chain, when the agent must act inside your own systems in ways a platform’s integrations cannot, or when latency and voice quality are part of your product. Many teams start on a platform and move once the numbers above favor it; designing prompts and tools to be portable makes that move cheaper. The same trade-off applies to text agents, and our guide to build vs buy AI agents runs it over three years.

Platform or custom, in short
Hosted platformVapi, Retell
  • Under about 10,000 minutes a month
  • A standard call flow
  • Testing whether callers accept an AI agent
  • Nobody to run voice infrastructure
Custom stackLiveKit, Pipecat
  • High or rising volume
  • Control over every vendor in the compliance chain
  • Actions inside your own systems
  • Latency and voice quality are part of your product

Is an AI voice agent worth it? Compare with missed calls

The right comparison is not the agent’s cost against a receptionist’s salary; it is the agent’s cost against the revenue lost to calls nobody answers. A business taking 60 calls a day at three minutes each uses about 5,400 minutes a month, which costs roughly $440 to $800 a month in platform usage at the list prices above.

Then estimate what missed calls cost you: calls a month, times the share you miss, times the share of those callers who would have bought, times the value of a customer. We walk through that model in what missed calls really cost, and you can run it with your own numbers in the AI voice agent ROI calculator. If the lost revenue is a multiple of the agent’s monthly cost, the case is easy at any of the prices in this guide.

How to lower the per-minute cost of a voice agent

The biggest savings come from the model and the voice, not from haggling over the platform fee.

  • Use the smallest model that passes your call tests, and keep the system prompt stable so it can be cached.
  • Pick a standard voice unless the voice is part of your brand; premium voices can cost several times more per minute.
  • Keep calls short with clear flows and fast handoff to a person when the agent cannot help.
  • Route by intent, so simple calls never reach the expensive model.
  • Recheck list prices before you commit. This guide uses prices checked on October 18, 2026, and vendors update them.

If you want these numbers worked out for your own call volume and systems, our AI voice agents team can model both routes in a free audit before you commit to either.

Related service
Want this applied to your business?
Free 45-min AI audit with a senior architect.
Book the audit
FAQ

Common questions.

How much does an AI voice agent cost per minute?

At list prices checked October 18, 2026, a hosted platform costs about $0.08 to $0.13 per minute on Vapi and $0.11 to $0.15 on Retell once you add speech-to-text, the language model, the voice and telephony. A custom stack built on Twilio, Deepgram and Claude costs about $0.04 per minute in usage, before hosting and the build.

How is AI voice agent pricing billed: per minute, per second or per month?

Mostly per minute of call time, with monthly fees on top. Vapi charges $0.05 a minute for hosting and passes the transcriber, model and voice through at provider cost. Retell prices each component per minute and bills to the second, including silence and hold (checked October 19, 2026). Plan fees, phone numbers and concurrent calls above the included number are billed monthly. A custom stack pays each vendor in its own unit: per minute for telephony and speech-to-text, per token for the model and per 1,000 characters for the voice.

Is per-call AI voice agent pricing cheaper than per-minute pricing?

Only for long calls. Ringly charges $1.99 per conversation after a $99 monthly minimum that covers the first 50 (checked November 10, 2026), which works out to about $0.66 a minute on a three-minute call. Against an all-in platform rate of about $0.11 a minute, per-call pricing only wins once calls average around 18 minutes. What it buys is a predictable bill and a finished receptionist product, not a platform you build on.

How much does it cost to build a custom AI voice agent?

A custom voice agent from Apptycoons costs $25k–$120k as a fixed-price build over 6–16 weeks. The quote comes at the end of a $5,000 two-week Discovery Sprint and depends on the systems the agent books into or updates, how many call flows it handles, and compliance needs such as HIPAA.

Is a platform like Vapi or Retell cheaper than a custom voice agent?

Below roughly 11,000 minutes a month, yes, on list prices checked October 18, 2026. A platform has no build to pay off and its per-minute fee covers infrastructure you would otherwise run yourself. Above roughly 31,000 minutes a month, a custom stack is cheaper against every platform configuration in our model. In between it depends on which platform and model you compare against.

What does an AI receptionist cost compared with a human?

At 5,400 minutes a month, about 60 calls a day at three minutes each, platform usage costs roughly $440 to $800 a month at list prices. A person covers business hours; the agent answers every call, including overlapping calls and after hours. So weigh the bill against the revenue lost to calls nobody picks up, not against a salary alone.

Do HIPAA-compliant voice agents cost more?

Usually, yes. Every vendor that handles patient data on the call has to sign a business associate agreement. Vapi lists HIPAA-eligible data handling as a $2,000 a month add-on, Retell’s plan comparison lists HIPAA and a BAA on pay-as-you-go as well as enterprise (a custom BAA is enterprise-only), and Twilio requires its Security or Enterprise Edition before it signs a BAA.

What are the line items in a voice agent’s per-minute cost?

Five: telephony to carry the call, speech-to-text to transcribe the caller, the language model that decides what to say, text-to-speech to speak the reply, and the platform or orchestration fee that ties them together. Platforms bundle these into one rate; a custom stack pays each vendor directly.

Do AI voice agents need consent to make outbound calls?

In the US, yes for most outbound calls. The FCC’s February 2024 declaratory ruling (FCC 24-17) confirmed that AI-generated voices count as an artificial or prerecorded voice under the TCPA, so calls that use them need the called party’s prior express consent unless an emergency purpose or exemption applies. An agent that only answers inbound calls does not face this rule. This is not legal advice.

Do Vapi or Retell charge a monthly fee on top of per-minute rates?

Neither charges a subscription on its entry plan, though both list fixed monthly costs. As of October 18, 2026, Vapi’s prepaid No Success Package costs $0 to start, Core costs $29 a month and Pro has a $999 a month minimum; HIPAA is a $2,000 a month add-on. Retell bills usage per minute and adds $2 a month per Retell phone number and $8 a month for each concurrent call above 20.