---
title: "Speko vs Retell AI"
description: "A sourced comparison of Speko (hosted phone agents, Router and Gateway) with Retell AI across product scope, telephony, routing, runtime ownership, observability and pricing."
canonical: "https://speko.ai/speko-vs-retell"
last-updated: "2026-11-03"
---

> ## Speko page index
> The complete index of every page on this site is at: https://speko.ai/llms.txt
> Read it before exploring further. It lists the exact Markdown URL for every canonical HTML page.

# Speko vs Retell AI

Comparison

Speko offers hosted phone agents (Speko Agents), a hosted model router (Speko Router) and an open customer-side Gateway; Retell AI is a hosted phone-agent platform that runs the agent for you, so the useful comparison is which boundary your team wants to own.

Short answer

Retell AI and Speko Agents are both hosted phone-agent platforms. Speko Agents bills $0.09 per minute all-in plus carriage on Speko numbers, with phone numbers, inbound and outbound calls, SMS and transfers included; Retell prices voice infrastructure, TTS, LLM, telephony and optional features as components. Choose Speko Router + Gateway instead when your own runtime owns the conversation loop and model selection is the boundary to make interchangeable.

Author

Speko engineering

Reviewed by

Speko engineering

Published

August 7, 2026 Last reviewed November 3, 2026

## How do Speko and Retell AI compare?

Retell AI category, scope and billing statements were checked against the official sources below on July 31, 2026. Speko Agents cells were checked against the Speko docs and pricing page on November 3, 2026. The table compares product boundaries, not overall quality. A dash means this site publishes no sourced statement for that cell.

| Dimension | Speko Agents (hosted) Hosted phone agents | Speko Router + Gateway Model layer, your runtime | Retell AI |
| --- | --- | --- | --- |
| Category | Hosted voice agent platform with phone numbers, calls and SMS. | Voice model router. | Hosted phone-agent platform. |
| Scope | The whole agent: prompt, voice, tools, knowledge and the phone number live in the Speko console and API. Speko runs the conversation loop. | Compatible model calls for STT, LLM and TTS. It does not run the conversation loop. | Managed phone agents assembled from Retell-supported infrastructure and model components. |
| Telephony | Included. Inbound and outbound phone calls on a Speko number or on a number you already own (SIP trunk import); SMS on Speko numbers; blind and warm transfers to a human. | Not included in the Router or Gateway boundary. Telephony and WebRTC remain with the application runtime. | Published options include Retell carrier service and custom SIP, priced as separate components. |
| Model routing | Measured selection and pre-response failover across STT, LLM and TTS. An agent can also pin providers, or run on a speech-to-speech model: Grok Voice 3.0 at the same per-minute rate, or GPT-Live, which adds its reasoning model's tokens at the provider's published rate plus 5%. | The application can pin a model or request measured automatic selection from eligible routes. | The agent uses model and infrastructure choices supported by the Retell configuration surface. |
| Runtime ownership | Speko operates the agent runtime and the phone legs, records the call, persists transcripts and events, and finalizes a call report after hangup. The customer configures the agent. | Speko operates the routing API; the customer keeps LiveKit, Pipecat or a custom runtime. | Retell operates the phone-agent platform and runtime; the customer configures the agent. |
| Observability and failover | Lifecycle events, a call report and a recording per call; signed workspace webhooks; call lists of up to 1,000 rows with a call window. | Route, reason, failover-count and first-byte headers. Failover is pre-response only, not mid-stream. | Call operations and logs are platform-managed within the supported component boundary. |
| Benchmark transparency | An agent's default stack is computed from the same benchmark dataset that feeds the public boards at speko.ai/benchmarks; the picks are written into the agent at create time. | Public voice benchmark boards at speko.ai/benchmarks; automatic selection ranks over the published snapshot. | - |
| Pricing model | $0.09 per minute all-in for STT, LLM and TTS, billed from answer to hang-up; unanswered calls cost nothing. Calls on a Speko number add the carrier's charge plus 20%, at least $0.015 per minute outbound and $0.01 per minute inbound; calls over your own SIP trunk add nothing. A Speko number is $1 setup and $2 a month. $10 signup credit. Prices as of November 3, 2026. | $0.09 per minute on Speko infrastructure for STT, LLM and TTS; phone calls on Speko numbers add the carrier's charge for the call, including its taxes and fees, plus 20%. Current terms are kept on the pricing page. | Component pricing: voice infrastructure plus selected TTS, LLM, telephony and optional features. |

Retell AI sources:

[Retell AI pricing](https://www.retellai.com/pricing) [Retell AI documentation](https://docs.retellai.com/)

Checked 2026-07-31

## Published per-minute prices

Data as of November 2026.

| Platform | Per minute | Included | Telephony | Monthly fee | Source |
| --- | --- | --- | --- | --- | --- |
| Speko our product | $0.09 per minute, all-in | Speech-to-text, LLM and text-to-speech on Speko infrastructure. Billed per second from answer to hang-up; unanswered calls cost nothing. | Calls on Speko numbers add the carrier's charge for the call, including its taxes and fees, plus 20%, at least $0.01 per minute inbound and $0.015 per minute outbound. Calls over your own SIP trunk add nothing. | None. A Speko phone number is $1 setup and $2 a month for most numbers. | [speko.ai/pricing](https://speko.ai/pricing) |
| Vapi | $0.05 per minute Vapi hosting, pre-paid | Vapi hosting. Speech-to-text, LLM and text-to-speech are passed through at the provider's cost on top. | Transport is charged by the carrier, not Vapi; Vapi's own SIP transport is listed as free. | None on the usage-only plan (1 included number, 4 concurrent calls). Core $29 a month; Pro $999 a month minimum; HIPAA $2,000 a month. | [vapi.ai/pricing](https://vapi.ai/pricing) |
| Retell AI | $0.07 to 0.31 per minute for AI voice agents | Retell voice infrastructure at $0.055 per minute plus the selected text-to-speech and LLM components, each priced per minute. | Retell-provided numbers $0.015 per minute in the US; own SIP trunk or custom telephony at no charge from Retell. | None. A Retell phone number is $2 a month; 20 concurrent calls included, $8 per extra concurrent call a month. | [retellai.com/pricing](https://www.retellai.com/pricing) |
| Bland AI | $0.14 per minute on Start; $0.12 per minute on Build | LLM, speech-to-text and text-to-speech, with no token charges. Transfer time bills $0.05 per minute on Start and $0.04 per minute on Build. | Billed separately: your own carrier, or Bland's built-in telephony at pass-through cost. | $0 platform fee on Start (an inbound number included); $299 a month platform fee on Build. | [bland.ai/pricing](https://www.bland.ai/pricing) |
| Synthflow | No per-minute rate published; annual enterprise contract | Scoped per contract. | Native telephony or SIP trunking, scoped per contract. | Quote only. No free plan or trial. | [synthflow.ai/pricing](https://synthflow.ai/pricing) |

## When to choose Retell AI

- The requirement is a managed phone-agent product rather than a standalone model-routing API.
- Telephony and agent operations should live in one hosted workflow.
- A component-based configuration and pricing surface fits how the team buys.

## When to choose Speko

- Speko Agents runs the whole phone agent: the prompt, the voice, its tools, its phone number, SMS and blind or warm transfers live in the console and API, at $0.09 per minute plus carriage on Speko numbers.
- Automatic routing considers only models in the published benchmark snapshot. If the request constraints leave no eligible candidate, the router returns no_candidate rather than selecting outside that snapshot.
- The route is decided when your session opens, not while you speak.
- A preview endpoint returns the provider Speko would pick for a language and objective, with its runners-up, before you send a request.

## Method and provenance

Every Retell AI cell is copied from the sourced comparison this site maintains on the [voice agent platforms page](https://speko.ai/voice-agent-platforms), where each statement carries its official source and check date. Speko Router + Gateway cells are verified against the public contracts and claim ledger this site builds from; Speko Agents cells against [Build a phone agent (Speko docs)](https://docs.speko.ai/guides/phone-agents) and the pricing page. The benchmark snapshot behind routing is published at [speko.ai/benchmarks/data.json](https://speko.ai/benchmarks/data.json).

Choose the layer first. Compare products inside that boundary second.

Related pages

- [Voice agent platforms, runtimes and model routers](https://speko.ai/voice-agent-platforms)
- [Speko vs Vapi](https://speko.ai/speko-vs-vapi)
- [Speko vs Bland AI](https://speko.ai/speko-vs-bland)
- [Vapi alternatives](https://speko.ai/vapi-alternatives)
- [Current model catalog](https://speko.ai/models)
- [Speko pricing](https://speko.ai/pricing)