At a glance
- Category
- Voice AI agents
- Tags
- Free trial, HIPAA
For developers
- Works with
- Make, Twilio, Vonage, Go High Level, n8n, HubSpot and 24 more
Retell AI in our research
All researchStudy
AI Tool Pricing Index, November 2026
Pricing
Usage-based
43 of the 361 tools we checked bill by usage.
Edition price, checked 14 November 2026.
Pricing pages of the AI tools, November 2026 editionas of n = 361 tools checked
What is Retell AI?
Retell AI is a voice automation platform for support, operations, sales, healthcare, and developers that handles phone calls with low-latency orchestration and natural turn-taking. It includes Call Transfer, Book Appointments, Knowledge Base, Navigate IVR, Batch Call, and real-time transcription and summarization. Retell integrates with Twilio, HubSpot, OpenAI, and Cal.com, and serves 3,000+ businesses. Plans run Pay as you go Free and Enterprise Plan custom.
A hosted voice agent stack for contact center calls
Retell runs speech recognition, the language model, text to speech and telephony as one hosted service, so a team configures an agent instead of wiring providers together. The homepage describes it as a way for "contact center teams [to] build, test, and deploy AI phone agents that answer and place calls, qualify leads, book appointments, and transfer to a human when needed."
The features it leads with are contact center mechanics: warm transfers where the agent briefs the human before connecting the caller, IVR navigation that works phone trees and enters extensions, deterministic steps for disclosures and identity checks mixed with free conversation, code execution and actions during a live call, and persistent memory that keeps "a running brief on every contact across calls and texts." Outbound campaigns add Do Not Call screening, local area code matching and automated retries. The homepage claims a response time of about 600 milliseconds. The product menu also lists AI receptionist and answering service use cases, but what is sold is the platform to build those agents, not a finished service.
The per-minute rate is a sum of components and add-ons
There is no platform fee. The pricing page quotes voice agents at "$0.07-$0.31 / min" on pay as you go and chat agents per AI message, for example $0.01 per message on GPT 6.1 Terra. New accounts get $10 in free credits, and Enterprise is "Custom Pricing" for both voice and chat.
The minute rate is built from parts: Retell's voice infrastructure, the chosen text to speech voice (ElevenLabs v3 and v4 run $0.10 a minute), the chosen LLM, and telephony, with "No Charge for Sip Trunking/Custom Telephony." Add-ons stack on top: PII removal at +$0.01 a minute, AI quality assurance at $0.10 a minute after the first 100 free minutes, branded caller ID at +$0.10 per outbound call, and a concurrency burst at +$0.10 a minute that lets calls above the limit connect instead of being rejected. SMS is $0.01 per message.
Monthly items sit beside usage: $2 per Retell phone number, $10 per verified phone number, $20 for Retell SMS, $8 per concurrent call past the 20 included free, and $8 per knowledge base past a free allowance of 10. Two LLM billing adjustments appear as their own invoice lines: calls under 10 seconds that open with a dynamic greeting bill as 10 seconds, and prompts averaging over 4,000 tokens bill proportionally more time, so "a 4,800-token prompt is 20% over, so the call bills 20% more time." The FAQ says calls are tracked to the nearest second, silence and hold are billed, failed calls are not, and the agent fee stops once a call is transferred. A calculator on the same page estimates a monthly bill.
The pricing page says no feature gating, then gates features
The pricing page states "No platform fees, no feature gating. Every plan gets full access to the entire platform." The Enterprise card on the same page lists what pay as you go does not get: "Custom MSA/DPA terms", "Role-based Access Control", "Custom SSO", a dedicated stable server and dedicated implementation support as add-ons, and "24/7 Omnichannel Support with Dedicated Portal." Pay as you go is limited to community and email support.
Enterprise concurrency is described three ways on that page. The plan card says "High Cap on Concurrent Calls", the comparison table says "No Cap on Concurrent Calls", and the FAQ says "Enterprise plans offer custom concurrency starting at 50+ calls."
The cost calculator also lags the rate table. The table marks GPT 6.1 Terra, GPT 6 Sol and Claude 5.5 Sonnet as recommended, while the calculator's LLM picker starts at GPT 5.6 Terra and offers none of the three, so an estimate cannot be built on the models Retell recommends. Retell's own model, Tell 1, is marked Beta in the rate table and is also missing from the picker.
REST API, SDKs and a hosted MCP server
"Webhooks and API Access" are listed on the pay as you go plan. The SDK docs describe official Node.js and Python clients, both published as retell-sdk, with rate limits set per organization and route (call creation at 1000 per 10 seconds). The same page warns that pinning an old version does not preserve old behavior: "the API changes on the server, so an old client stops matching what the API accepts."
The MCP server is remote, at mcp.retellai.com over Streamable HTTP with a Bearer API key, and covers agents, calls, phone numbers, knowledge bases, voices, chats, testing and alerts. Instead of one tool per endpoint it exposes three: list_api_endpoints, get_api_endpoint_schema and invoke_api_endpoint. The docs state the cost of that design: if a client always allows invoke_api_endpoint, "you allow every endpoint your API key can reach, including delete and publish." Separate Claude and ChatGPT connectors use a Retell login and in-chat widgets instead of an API key.
Compliance claims are broad, with contract terms on Enterprise
The homepage states Retell is "HIPAA and GDPR compliant, and SOC 2 Type 1 and Type 2 certified", says a BAA can be signed for healthcare workloads, and points to a Trust Center for the reports. ISO 27001 appears among its badges. It also describes user-defined PII redaction in calls and transcripts and configurable roles with module-level permissions. The pricing page words the same position more cautiously, "SOC 2 certified, HIPAA-ready", and places custom MSA, DPA and BAA terms, SSO and role-based access control on Enterprise.
A small, profitable team that started at Y Combinator
The About page dates the company to Y Combinator acceptance in November 2023 with a team of five, a public launch on Hacker News and Product Hunt in February 2024, and a seed round in May 2024 "backed by Alt Capital and an amazing group of angel investors." It says Retell became fully profitable with a team of 25 and that "More than 4,000 customers now run 100 million-plus calls a month on Retell." It names Bing Wu as CEO, Zexia Zhang as CTO, Todd Li as President, Weijia Yu as COO and Evie Wang as CMO, and lists angels including Aaron Levie and Michael Seibel. The revenue and customer figures are the company's own, and no independent audit of them is published.
Where it fits, and where it does not
It fits teams with real call volume and someone able to tune prompts, model choice, voice provider and add-ons, because each of those moves the per-minute cost. A buyer who wants one flat monthly fee, or a finished answering service that needs no configuration, gets neither: the entry is pay as you go with a one-time $10 credit, and SSO, role-based access, custom contract terms and round-the-clock support sit on Enterprise behind a sales conversation.
How much does Retell AI cost?
Prices excluding tax, as Retell AI publishes them.
Pay as you go
Free
- Go live in minutes with $10 in free credits. Full platform access, no commitments. Just build, test and pay as you scale.
- $0.002+/ msg for AI Chat Agents
- Full Platform Access
- Pre-built Agent Templates
- Call Analytics and Transcripts
- Simulation Testing
- Webhooks and API Access
- 20 Free Concurrent Calls
- Community and Email Support
Enterprise Plan
Custom Pricing
- Dedicated Implementation Support (add-on)
- Custom MSA/DPA terms
- Role-based Access Control
- Dedicated Stable Server
- Custom SSO
- High Cap on Concurrent Calls
- 24/7 Omnichannel Support with Dedicated Portal
Where these prices come from
retellai.com/pricing, checked
All plans and limits
| Plan | Price | Concurrent calls |
|---|---|---|
| Enterprise Plan | Price on request | No Cap on |
| Pay as you go | From $0.07 per minute | 20 Free |
Read from retellai.com/pricing.
For AI agents
- MCP server
- YesChecked by us,
- API
- Yesretellai.com,
What we checked on Retell AI
What the price counts
Retell charges its pay-as-you-go voice service by consumed minutes at a displayed per-minute range.
$0.07-$0.31 / min for AI Voice Agents $0.002+ / msg for AI Chat Agents
retellai.com/pricingChecked
Free to use
Self-serve signup with no sales call: pay-as-you-go has no platform fee and new accounts receive $10 in free credits, though ongoing use is metered and billed.
Go live in minutes with $10 in free credits. Full platform access, no commitments. Just build, test, and pay as you scale.
retellai.com/pricingChecked
Public API
A documented REST API with Node.js and Python SDKs covers agents, calls, phone numbers and chats, and it is listed as included on the entry pay-as-you-go plan.
Webhooks and API Access
retellai.com/pricingChecked
MCP server
Its MCP server at mcp.retellai.com answers, 3 tools
retell-sdk
mcp.retellai.com/mcpChecked by us,
Open source
8 public repositories on GitHub, none of them the product's code
On the page: github.com/RetellAI repositories
github.com/RetellAIChecked by us,
Retell AI alternatives
Questions about Retell AI
- How much does Retell AI cost?
- Retell AI costs $0.07 per minute. From retellai.com/pricing, checked 14 November 2026.
- Does Retell AI have an API or an MCP server?
- Retell AI has an API. From retellai.com/pricing, checked 4 November 2026. Retell AI has an MCP server. We checked mcp.retellai.com/mcp ourselves on 15 November 2026.
- What have we checked on Retell AI?
- 4 facts, under these headings: What it costs to start, how you build with it.
- How do we check Retell AI?
- We hold 4 facts on Retell AI, quoted from 4 pages. Facts last checked 15 November 2026. Price last checked 14 November 2026. Last change found 15 November 2026.
Sources and changes
Last full check
Sources3
What changed3
- What it does
- What it does
- Free tier
Retell AI price history
Plans for Retell AI started at $0.07 per minute on both of our reads, from 30 October 2026 to 14 November 2026.
| Price | Plan | First seen | Last seen | Readings | Source |
|---|---|---|---|---|---|
| $0.07 per minute | 30 October 2026 | 14 November 2026 | 2 | Pricing page, read 30 October 2026 |
From our research, read twice since 30 October 2026. 12 more of our checks could not pick this price out of the page and are left out. A date is when we read the page, not when the price changed.
Is Retell AI yours?
Free, and the page then carries the verified owner mark. You confirm it with an email at retellai.com. Claim this listing
Report an error if a fact is wrong.