Retell AI Pricing (2026): Per Minute Costs and Real Bills
Every Retell AI line item, from voice infra to LLMs, voices, telephony and add ons, plus the monthly bill at five volumes and how it compares, as of October 2026.
By , research by the BestBots team · Updated Oct 19, 2026
Fact checked against each company's published pricing, documentation and verified user ratings. How we score
Retell AI costs $0.055 a minute for voice infra in October 2026, plus the LLM, voice and telephony you choose, which Retell says lands at $0.07 to $0.31 a minute all in. A budget stack is about $0.0946 a minute. The first 20 concurrent calls are free, HIPAA has no surcharge, and new accounts get $10 in credits.
- Starting price
- $0.055 per minute
- Pricing model
- per minute, pay as you go
- Free plan
- No
- Free trial
- $10 in free credits
- Per minute cost
- $0.055 voice infra plus LLM ($0.0016 to $0.32), voice ($0.015 to $0.10) and telephony ($0.015 US Twilio or Telnyx)
- Latency
- About 600 ms, per Retell's homepage
- Telephony
- Retell numbers via Twilio, Vonage, or your own SIP trunk at no Retell charge
- Concurrency
- 20 concurrent calls free, then $8 per line per month
- Languages
- Multilingual agents supported; the dashboard lists current languages
- HIPAA
- Yes, listed as HIPAA ready
- SOC 2
- SOC 2 Type II
The price list
Per minute components
| Component | Options | Price per minute |
|---|---|---|
| Voice infra | Retell, includes speech to text | $0.055 |
| LLM, budget | GPT 5 nano $0.0016, Claude 5.5 Haiku or GPT 6 Luna $0.0032, Gemini 3.5 Flash Lite $0.0096 | $0.0016 to $0.0096 |
| LLM, recommended | GPT 5.6 Terra, Claude 5 Sonnet, Claude 5.5 Sonnet | $0.064 |
| LLM, premium | GPT 5.4 $0.080, GPT 5.5 $0.16, GPT 6 Astra $0.32 | $0.08 to $0.32 |
| LLM fast tier | Most models | Double the standard rate |
| Voice | Retell, Minimax, Fish, Cartesia, OpenAI, Inworld | $0.015 |
| Voice | ElevenLabs Flash / Multilingual v2 / v3 Conversational / v3 and v4 | $0.040 / $0.06 / $0.08 / $0.10 |
| Telephony | Retell numbers on US Twilio or Telnyx | $0.015 |
| Telephony | SIP trunking or custom telephony | No charge |
| Speech to speech | GPT Realtime, GPT Realtime 1.5 | $0.345 (replaces LLM and voice) |
Add ons and monthly items
| Item | Price |
|---|---|
| Concurrency | 20 free, then $8 per line per month |
| Phone numbers | $2 per month (verified numbers $10) |
| Knowledge bases | 10 free, then $8 each per month, plus $0.005 per minute when used |
| Advanced denoising | $0.005 per minute |
| Safety guardrails | $0.005 per minute |
| PII removal | $0.01 per minute |
| AI quality assurance | First 100 minutes free, then $0.10 per minute |
| Batch calls | $0.005 per dial |
| Branded calls | $0.10 per outbound call |
| SMS | $0.01 per message; Retell SMS $20 per month |
Enterprise is custom, with dedicated servers, no concurrency cap, HIPAA BAA, SSO, custom MSA, DPA and data retention, and 24/7 support.
Three stacks priced
| Stack | Infra | LLM | Voice | Telephony | Per minute |
|---|---|---|---|---|---|
| Lowest published | $0.055 | GPT 5 nano $0.0016 | Retell $0.015 | $0.015 | $0.0866 |
| Budget | $0.055 | Gemini 3.5 Flash Lite $0.0096 | Retell $0.015 | $0.015 | $0.0946 |
| Recommended | $0.055 | GPT 5.6 Terra $0.064 | Retell $0.015 | $0.015 | $0.149 |
| Premium voice | $0.055 | GPT 5.6 Terra $0.064 | ElevenLabs v4 $0.10 | $0.015 | $0.234 |
The cheapest models are fine for scripted booking and FAQ calls; open ended support usually needs a recommended model. Add a knowledge base ($0.005) and guardrails ($0.005) and the recommended stack is $0.159.
Worked examples
| Use case | Minutes a month | Stack | Concurrency | Math | Monthly bill |
|---|---|---|---|---|---|
| Prototype | 500 | Budget | 2 | 500 x $0.0946 | $47.30 |
| Small business line | 3,000 | Recommended | 4 | 3,000 x $0.149 | $447 |
| Busy inbound line | 10,000 | Budget + KB | 10 | 10,000 x $0.0996 | $996 |
| Support center | 30,000 | Recommended + KB + guardrails | 30 | 30,000 x $0.159 + 10 lines x $8 | $4,850 |
| Healthcare line | 30,000 | Budget + PII removal | 20 | 30,000 x $0.1046 | $3,138 |
Phone numbers add $2 a month each. Healthcare carries no HIPAA surcharge on Retell's published pricing, which is the biggest difference from Vapi's $2,000 a month add on.
Hidden costs to plan for
- The headline is partial. $0.055 is infra only; a real call is 1.6 to 5.6 times that.
- Model choice swings the bill. LLMs range 200x, from $0.0016 to $0.32 a minute.
- Premium voices. ElevenLabs v3 and v4 cost almost twice the infra rate.
- Stacked add ons. QA at $0.10 a minute doubles a budget call.
- Fast tier. Lower latency LLM tiers cost double.
How Retell compares per minute
| Platform | Headline | Budget all in | Free concurrency | HIPAA |
|---|---|---|---|---|
| Telnyx Voice AI | $0.05 incl. STT and TTS | About $0.06 | 500 | Included |
| Vapi | $0.05 hosting | About $0.0818 | 4 | $2,000 a month |
| ElevenLabs Agents | $0.08 after plan minutes | $0.08 plus LLM and carrier | 4 to 40 | Enterprise BAA |
| Retell AI | $0.055 infra | About $0.0946 | 20 | Included |
| Bland AI | $0.11 to $0.14 all in | Same | 10 | Listed |
Retell sits in the middle on per minute cost and near the top on what you get for free: 20 lines, HIPAA and a full price sheet. See the AI Price Index and the voice agent cost per minute guide.
How to keep the bill down
- Test the cheapest LLM that passes your scripts before defaulting to a recommended model.
- Use SIP trunking or your own carrier; Retell charges nothing for custom telephony.
- Stay within 20 concurrent lines until volume justifies $8 a line.
- Turn on add ons per agent, not globally.
Is Retell worth the price?
Yes, for most businesses building a phone agent: every price is public, 20 lines and HIPAA are included, and the no code builder plus API serves both operators and developers. If cost per minute at very high volume is the only metric, Telnyx is cheaper. Read the full Retell AI review, compare Retell AI vs Vapi or see Retell AI alternatives.
Retell AI plans and pricing
| Plan | Price | What you get |
|---|---|---|
| Pay as you go | $0 per month | $10 free credits, 20 concurrent calls, $0.055 per minute voice infra plus LLM, voice and telephony |
| Enterprise | Custom | Custom pricing and terms |
Frequently asked questions
How much does Retell AI cost per minute?
Retell charges $0.055 a minute for voice infra, which includes speech to text, plus the LLM ($0.0016 to $0.32 a minute), the voice ($0.015 to $0.10) and telephony ($0.015 a minute on US Twilio or Telnyx). Retell's pricing page quotes $0.07 to $0.31 a minute all in.
What is the cheapest Retell AI setup?
On published rates, GPT 5 nano at $0.0016 a minute with a $0.015 voice and $0.015 telephony comes to about $0.0866 a minute. A more common budget stack, Gemini 3.5 Flash Lite at $0.0096 with a Retell voice and US Twilio, is about $0.0946.
Does Retell AI charge for concurrency?
The first 20 concurrent calls are free. Each additional concurrent line costs $8 a month.
Is Retell AI HIPAA compliant, and does it cost extra?
Retell lists itself as HIPAA ready with SOC 2 Type II, and HIPAA does not carry a published surcharge on pay as you go. Custom BAA, SSO and custom data retention are listed on Enterprise.
Does Retell AI have a free trial?
Pay as you go starts with $10 in free credits and no monthly fee.
What add ons does Retell charge for?
Per minute add ons include knowledge base use ($0.005), advanced denoising ($0.005), safety guardrails ($0.005), PII removal ($0.01) and AI quality assurance ($0.10 after 100 free minutes). Batch calls are $0.005 per dial, branded calls $0.10 per outbound call and SMS $0.01 per message.
Is Retell cheaper than Vapi?
Not per minute on the cheapest components: Vapi's budget stack is about $0.0818 against Retell's $0.0946. Retell is usually cheaper in total once you need more than 4 concurrent lines or HIPAA, which Vapi charges $10 per line and $2,000 a month for.