Major Takeaways
Who should pick Retell AI over Vapi? Teams that prioritize documented enterprise compliance features, pre-built integrations with CRMs and calendars, and a platform optimized for production-scale customer-service and sales use cases.
Who should pick Vapi over Retell AI? Developers building custom voice products who need maximum flexibility in LLM choice, voice provider selection, and telephony configuration, and who value transparent per-minute pricing without platform fees.
How do their pricing models differ? Retell AI reportedly charges per minute plus platform fees; Vapi publishes a per-minute rate that includes LLM, TTS, and STT costs, with no separate platform subscription for most tiers.
If you care about X, pick Y
| If you care about | Pick | Why |
|---|---|---|
| Transparent published pricing | Vapi | Vapi lists per-minute rates on its pricing page; Retell AI requires contact for custom quotes. |
| Enterprise compliance documentation | Retell AI | Retell AI publishes SOC 2 Type II and HIPAA compliance statements; Vapi's compliance posture is less documented publicly. |
| Maximum LLM flexibility | Vapi | Vapi supports bring-your-own LLM keys and multiple providers; Retell AI's model options are less extensively documented. |
| Pre-built CRM integrations | Retell AI | Retell AI documents native integrations with Salesforce, HubSpot, and other CRMs; Vapi requires custom webhook builds. |
| Low-latency voice interactions | Either | Both platforms claim sub-second latency; published benchmarks are vendor-stated and not independently verified. |
| Outbound campaign tooling | Retell AI | Retell AI documents outbound dialing and campaign management; Vapi's outbound features are less extensively detailed in public docs. |
How we evaluated
This comparison draws on vendor documentation from Retell AI's website and Vapi's documentation, pricing pages accessed in October 2026, and third-party reviews on G2 and TrustRadius. We assessed build experience, telephony options, LLM and voice provider flexibility, published compliance statements, pricing transparency, and documented use-case fit.
No test calls were placed. No agents were built. No latency measurements were conducted.
This is desk research based on publicly available information. Last updated: October 2026.
What each platform does
Retell AI
Retell AI is an API-first AI voice agent platform designed for engineering teams and enterprises automating customer service, sales qualification, and appointment booking. Users build agents via API or a low-code dashboard, configure conversation flows, and deploy to inbound or outbound telephony.
The platform reportedly supports warm transfers to human agents, function calling for CRM lookups and calendar bookings, and integrations with Salesforce, HubSpot, and other enterprise systems. Retell AI states it has customers in healthcare, financial services, and contact-center operations. The company publishes SOC 2 Type II and HIPAA compliance documentation, positioning itself for regulated industries.
Vapi
Vapi is a developer voice AI platform emphasizing low-latency conversational AI and maximum configurability. Developers build agents with Vapi's API and SDKs, selecting their own LLM provider (OpenAI, Anthropic, or bring-your-own), voice provider (ElevenLabs, PlayHT, Azure, or custom), and telephony setup (Vapi-provided numbers, Twilio, or SIP trunking).
Vapi reportedly supports inbound and outbound calls, real-time function calling, and webhook-based integrations. The platform is used by startups and technical teams building custom voice products, customer-support bots, and sales agents. Vapi publishes transparent per-minute pricing that bundles LLM, TTS, and STT costs, with no separate platform fee for most use cases.
Feature comparison
| Capability | Retell AI | Vapi |
|---|---|---|
| Build model | API-first with low-code dashboard | API-first, SDK-based |
| Inbound calls | ✓ | ✓ |
| Outbound calls / campaigns | ✓ Documented campaign management | ✓ API-driven, less campaign tooling |
| Telephony | Bundled numbers, bring-your-own carrier, SIP | Bundled numbers, Twilio, SIP trunking |
| LLM choice | Limited documentation on bring-your-own | ✓ OpenAI, Anthropic, bring-your-own keys |
| Voice / TTS provider choice | Vendor-hosted voices, limited cloning details | ✓ ElevenLabs, PlayHT, Azure, custom |
| Languages | Not extensively documented | Supports 29+ languages per TTS provider |
| Interruption handling | ✓ Documented turn-taking controls | ✓ Configurable interruption thresholds |
| Warm transfer / human handoff | ✓ | ✓ |
| Function calling / tool use | ✓ CRM lookups, calendar bookings | ✓ Real-time function calling via webhooks |
| CRM and calendar integrations | ✓ Native Salesforce, HubSpot, Calendly | Webhook-based, requires custom build |
| Call recording, transcripts, analytics | ✓ | ✓ |
| Compliance statements | SOC 2 Type II, HIPAA (published) | Not extensively documented publicly |
| Concurrency limits | Not publicly listed | Not publicly listed |
Pricing and cost per minute
Retell AI pricing
Retell AI does not publish list pricing on its website. Prospective customers must contact sales for a custom quote.
Based on third-party sources and vendor statements, Retell AI reportedly charges per minute of call time plus a platform fee or seat-based subscription, with pricing varying by volume, use case, and contract length. The per-minute rate may or may not include telephony, LLM inference, and TTS costs; this is clarified during the sales process.
Enterprise contracts reportedly include dedicated support, compliance documentation, and custom SLA terms. No free tier or self-serve trial is publicly advertised.
| Tier | Pricing model | Notes |
|---|---|---|
| Custom | Contact sales | Per-minute rate + platform fee; pricing varies by volume and contract |
Vapi pricing
Vapi publishes transparent per-minute pricing on its website. As of October 2026, Vapi's standard rate is $0.05 per minute, which reportedly includes LLM inference (OpenAI GPT-4 or equivalent), text-to-speech, and speech-to-text costs.
Telephony (inbound and outbound calling via Vapi-provided numbers or Twilio) is billed separately at carrier rates, typically $0.01–$0.02 per minute depending on region. There is no separate platform fee or seat charge for most use cases.
Vapi offers a free tier with limited monthly minutes for testing. Volume discounts are available for enterprise customers. Bring-your-own LLM or TTS keys may reduce the per-minute rate, as the customer pays their provider directly.
| Tier | Per-minute rate | Telephony | Notes |
|---|---|---|---|
| Free | $0.00 | Limited minutes | Testing and development |
| Standard | $0.05 | $0.01–$0.02 | Includes LLM, TTS, STT |
| Enterprise | Custom | Custom | Volume discounts, dedicated support |
Source: Vapi pricing page, accessed October 2026.
What a month of calls might cost
To estimate monthly spend, use this formula:
Total cost = (average call length in minutes × number of calls per month × all-in per-minute rate) + platform fees + telephony charges
For Vapi, if you assume an illustrative average call length of 3 minutes, 1,000 calls per month, and the standard $0.05 per minute rate plus $0.015 per minute for telephony, the calculation is:
- Call cost: 3 minutes × 1,000 calls × $0.05 = $150
- Telephony: 3 minutes × 1,000 calls × $0.015 = $45
- Total: $195 per month
For Retell AI, the same volume would depend on the negotiated per-minute rate and platform fee. If the per-minute rate were $0.08 (an illustrative assumption; actual rates vary) and a platform fee of $500 per month applied, the calculation might be:
- Call cost: 3 minutes × 1,000 calls × $0.08 = $240
- Platform fee: $500
- Total: $740 per month
These are illustrative examples only. Replace the call length, volume, and rates with your own numbers. Request detailed quotes from both vendors to confirm what the per-minute rate includes and whether platform fees, minimums, or overage charges apply.
What users say
Public review coverage for both platforms was limited at the time of writing. Developer-focused voice AI platforms often have thin representation on G2 and TrustRadius compared to broader SaaS categories. The themes below are drawn from available sources.
Retell AI
Public review volume for Retell AI is limited; the recurring themes below are drawn from the available G2 / TrustRadius coverage.
Users reportedly appreciate Retell AI's focus on enterprise compliance and pre-built integrations, which reduce engineering lift for teams deploying in regulated industries. Some reviewers note that the lack of published pricing and the requirement to engage sales for a quote can slow procurement cycles.
Documentation quality and API clarity receive mixed feedback. Some developers find the onboarding process smoother than expected. Others request more detailed examples for advanced use cases.
Vapi
Developers reportedly value Vapi's transparent pricing, flexible LLM and TTS provider options, and straightforward API design. The platform's emphasis on configurability is cited as a strength for teams building custom voice products.
Some users note that the lack of pre-built CRM integrations means more webhook development work compared to platforms with native connectors. A subset of reviewers mention that Vapi's documentation, while thorough, assumes a higher level of technical familiarity than some competitors.
Strengths and gaps
What Retell AI does well
Enterprise compliance posture. Retell AI publishes SOC 2 Type II and HIPAA compliance statements, making it a stronger fit for healthcare, financial services, and other regulated industries where documented compliance is a procurement requirement.
Pre-built integrations. Native connectors to Salesforce, HubSpot, Calendly, and other CRMs reduce engineering time for teams deploying voice agents in existing sales and support workflows.
Outbound campaign tooling. Retell AI documents outbound dialing and campaign management, positioning it for teams running high-volume outbound call programs.
Warm transfer and human handoff. The platform supports transferring calls to live agents, a critical feature for customer-service and sales use cases where escalation is common.
Where Retell AI has gaps
Retell AI does not publish list pricing. Every prospect must engage a sales process, which can slow evaluation and procurement for smaller teams or startups.
Retell AI's documentation does not extensively detail bring-your-own LLM or TTS options, which may limit customization for teams with specific model or voice requirements.
No publicly advertised free tier or self-serve trial is available, which raises the barrier to entry for developers who prefer to test before committing to a sales conversation.
Limited third-party review coverage makes it harder for buyers to assess real-world user sentiment and edge-case performance.
What Vapi does well
Transparent pricing. Vapi publishes per-minute rates on its website, with clear breakdowns of what the rate includes (LLM, TTS, STT) and what is billed separately (telephony). This transparency accelerates procurement and budgeting.
LLM and TTS flexibility. Vapi supports bring-your-own LLM keys (OpenAI, Anthropic, or custom) and multiple TTS providers (ElevenLabs, PlayHT, Azure), giving developers maximum control over model choice and voice quality.
Developer-first API design. The platform's API and SDK are reportedly straightforward and well-documented, with a focus on configurability and low-latency performance.
Free tier for testing. Vapi offers a free tier with limited monthly minutes, allowing developers to prototype and test agents before committing to paid usage.
Where Vapi has gaps
Vapi does not offer native CRM or calendar connectors. Integrations require custom webhook builds, which adds engineering time for teams without in-house developers.
Vapi's public compliance posture (SOC 2, HIPAA, GDPR) is less extensively documented than Retell AI's, which may be a barrier for regulated industries.
While Vapi supports outbound calling via API, it does not document the same level of campaign management and dialing features as Retell AI, which may limit fit for high-volume outbound programs.
Vapi's standard tier does not include dedicated support. Enterprise customers must negotiate custom contracts for SLA-backed assistance.
Which use cases fit each
| Use case | Stronger fit | Why |
|---|---|---|
| Inbound lead qualification and speed-to-lead | Either | Both platforms handle inbound calls with low latency; Retell AI's CRM integrations may reduce setup time, while Vapi's flexibility suits custom workflows. |
| Appointment booking | Retell AI | Native calendar integrations (Calendly, Google Calendar) reduce engineering lift for booking use cases. |
| After-hours reception | Either | Both support 24/7 inbound call handling; choice depends on integration and compliance requirements. |
| Outbound reminders and confirmations | Retell AI | Documented outbound campaign management makes Retell AI the stronger fit for high-volume reminder programs. |
| Outbound sales calling | Retell AI | Retell AI's outbound tooling and CRM integrations position it for sales teams running structured call campaigns. |
| Customer-support deflection | Retell AI | Warm transfer to human agents and native CRM connectors make Retell AI the better starting point for support deflection. |
| High-volume contact-centre overflow | Retell AI | Enterprise compliance documentation and support model suit contact-centre deployments in regulated industries. |
| Developer building a custom voice product | Vapi | Maximum LLM and TTS flexibility, transparent pricing, and a free tier make Vapi the stronger fit for custom product builds. |
Pick Retell AI if you...
- Need documented SOC 2 Type II or HIPAA compliance for a regulated industry deployment.
- Want native integrations with Salesforce, HubSpot, or Calendly to reduce engineering time.
- Are running high-volume outbound call campaigns and need documented campaign management features.
- Prefer a platform optimized for enterprise customer-service and sales use cases with dedicated support.
Pick Vapi if you...
- Value transparent per-minute pricing and want to avoid a sales process for initial evaluation.
- Need maximum flexibility in LLM choice, voice provider selection, and telephony configuration.
- Are building a custom voice product and want a developer-first API with a free tier for testing.
- Have in-house engineering resources to build webhook-based integrations and prefer configurability over pre-built connectors.
Note: if the calls you're automating are inbound sales conversations that need to sit alongside an outbound AI SDR in one platform, teams in that position typically also evaluate 11x's Julian.
Verdict
Retell AI is the stronger fit for enterprises and revenue teams deploying voice agents in regulated industries or high-volume sales and support workflows. The platform's published compliance documentation, native CRM and calendar integrations, and documented outbound campaign tooling reduce engineering lift and procurement friction for teams that need a production-ready solution with enterprise support.
The lack of published pricing and limited LLM flexibility may slow evaluation for smaller teams or developers who prioritize configurability.
Vapi is the better starting point for developers and technical teams building custom voice products or deploying agents in environments where LLM choice, voice provider selection, and transparent pricing are priorities. The platform's developer-first API, free tier, and bring-your-own-key model make it the more flexible option for teams with in-house engineering resources.
The absence of pre-built CRM integrations and less extensive compliance documentation may limit fit for regulated industries or teams without dedicated developers.
For teams evaluating both, build the same short test flow on each platform (e.g., a simple inbound qualification script with one function call and one warm transfer). Compare the all-in per-minute cost at your expected volume, including telephony and any platform fees. Confirm compliance documentation and SLA terms with each vendor if your use case requires it. Request detailed quotes to clarify what the per-minute rate includes and whether minimums or overage charges apply.
FAQ
Is Retell AI cheaper than Vapi?
Retell AI does not publish list pricing, so direct cost comparison requires a custom quote. Vapi's published rate of $0.05 per minute (plus telephony at $0.01–$0.02 per minute) is transparent and may be lower than Retell AI's negotiated rate for low-to-moderate volumes, but Retell AI's pricing may be competitive at enterprise scale. Platform fees and contract minimums can shift the total cost significantly; request detailed quotes from both vendors.
Which is easier to build with if I can't code?
Neither platform offers a true no-code builder. Both require API or SDK integration, which assumes developer resources. Retell AI's low-code dashboard may reduce some configuration complexity, but building and deploying an agent still requires technical familiarity. Teams without in-house developers should budget for external engineering support or consider platforms with visual builders.
Can I bring my own phone numbers or carrier to each?
Yes, both platforms support bring-your-own carrier and SIP trunking. Retell AI documents SIP integration for customers who want to use their existing telephony infrastructure. Vapi supports Twilio integration and SIP trunking, allowing customers to use their own numbers and carrier contracts. Confirm setup requirements and any additional fees with each vendor.
Which handles outbound calling campaigns better?
Retell AI is the stronger fit for outbound campaigns. The platform documents outbound dialing, campaign management, and list-based calling features designed for high-volume sales and reminder programs. Vapi supports outbound calling via API but does not document the same level of campaign tooling, which may require more custom development for teams running structured outbound programs.
Are Retell AI and Vapi HIPAA compliant?
Retell AI publishes a HIPAA compliance statement and offers Business Associate Agreements (BAAs) for healthcare customers. Vapi's HIPAA compliance posture is not extensively documented on its public website; teams deploying in healthcare should request compliance documentation and BAA terms directly from Vapi. Note that platform compliance does not guarantee the customer's deployment is compliant; compliance depends on how the customer configures call recording, data retention, and consent workflows.
Can either transfer a call to a human?
Yes, both platforms support warm transfer to human agents. Retell AI documents this feature as part of its core offering, designed for customer-service and sales use cases where escalation is common. Vapi also supports warm transfer via API, though the feature is less prominently documented. Confirm transfer latency, fallback behavior, and integration with your existing phone system with each vendor.
What do users complain about most?
Public review volume for both platforms is limited. For Retell AI, users reportedly cite the lack of published pricing and the requirement to engage sales as friction points. For Vapi, some developers note that the absence of pre-built CRM integrations means more webhook development work. Documentation quality and API clarity receive mixed feedback for both platforms, with some users finding onboarding smooth and others requesting more detailed examples for advanced use cases.
How does 11x's Julian compare to Retell AI and Vapi?
Julian is 11x's AI phone agent for inbound speed-to-lead and qualification, sold as part of the 11x platform alongside Alice (outbound AI SDR). Julian is a packaged sales agent rather than a general-purpose voice-building platform, so teams building custom voice products typically compare it less directly to Retell AI or Vapi. Julian is designed for revenue teams automating inbound sales conversations and routing qualified leads to human reps, with native integration to 11x's outbound motion. Teams evaluating Julian are usually choosing between a unified sales platform (11x) and a voice-building platform (Retell AI or Vapi) plus separate outbound tooling.
Can I test either platform before committing?
Vapi offers a free tier with limited monthly minutes, allowing developers to prototype and test agents before committing to paid usage. Retell AI does not publicly advertise a free tier or self-serve trial; prospective customers must contact sales to arrange a demo or pilot. Teams that prefer to test before engaging a sales process will find Vapi's free tier the lower-friction option.
Which platform has better documentation?
Both platforms reportedly have well-documented APIs, though user feedback is mixed. Vapi's documentation is cited as thorough but assumes a higher level of technical familiarity. Retell AI's documentation receives mixed feedback, with some developers finding onboarding smooth and others requesting more detailed examples for advanced use cases. Teams should review the public docs for each platform to assess fit for their technical skill level.
Last updated: October 2026.
