Cartesia vs Vapi: pricing, features and which to choose

Cartesia is a San Francisco startup founded in 2023 by Karan Goel, Albert Gu, Arjun Desai, Brandon Yang and Chris Ré, researchers from Stanford's AI lab. Gu co-created the S4 and Mamba state space model architectures, which process long sequences with less memory than transformers, and that research is the basis of the company's products. Cartesia launched its Sonic text-to-speech model in 2024 and raised a $64 million Series A led by Kleiner Perkins in March 2025, bringing total funding to $91 million, according to Fortune. Its homepage names ServiceNow, Retell, Blue Machines and Quora among customers. The company's pitch is speed: models that start speaking quickly enough for live phone conversation, sold through one API that covers speech generation, transcription and complete voice agents.

Developers reach Cartesia through a REST and WebSocket API or the browser playground. Sonic turns text into streaming audio, with controls for speed, emotion and pronunciation, and supports more than 40 languages. Instant voice cloning creates a voice from a few seconds of audio on paid plans, and professional cloning trains a closer copy from longer recordings on the Startup plan and above. A localisation feature gives a cloned voice a new accent. Ink transcribes live audio for agents that need to hear callers. Managed Agents, the newest product, hosts complete phone agents with telephony, so a team can define an agent's prompt and tools without wiring a separate orchestration layer. Cartesia also offers on-premises and on-device deployment for customers with data residency rules.

Cartesia's voice library and creator tools are thinner than ElevenLabs, which also offers dubbing, sound effects and music, so it is a weaker fit for video producers and audiobook narrators. Managed Agents are newer than Vapi or Retell and have fewer integrations. Credit-based billing means long-form narration gets expensive faster on the low tiers. ElevenLabs Flash competes directly on latency with a larger voice catalog. Deepgram's Aura voices are cheaper per minute but less expressive, and Hume, once a rival on expressive speech, is closing its TTS API in November 2026. Pick Cartesia when you are building a real-time voice agent or interactive app and the delay before the first syllable is the metric you care about, or when you need on-device or on-premises speech models.

Vapi is a San Francisco company that sells the plumbing behind AI phone agents. Jordan Dearsley and Nikhil Gupta, classmates from the University of Waterloo, went through Y Combinator with a productivity app called Superpowered. In 2023 Dearsley built an AI therapist to talk to on walks, found that startups wanted the low-latency voice layer underneath it, and the pair relaunched as Vapi in 2024. In May 2026 it raised a $50 million Series B led by Peak XV at about a $500 million valuation, bringing total funding to $72 million, according to TechCrunch. The company says it has handled more than 1 billion calls and has over 1 million developers on its self-serve platform. Enterprise customers include Amazon Ring, which routes all of its inbound support calls through Vapi, plus Intuit, New York Life and Kavak.

A Vapi agent is a configuration that wires together three parts: a transcriber that turns caller speech into text, a language model that decides what to say, and a voice that speaks the reply. You set these up in the dashboard or through the API, write a system prompt, and attach tools so the agent can book appointments, look up orders or transfer to a human. Vapi supplies phone numbers, or you bring your own through Twilio, Vonage or SIP. Squads let several specialised agents hand a call to each other. Testing and simulation features run scripted calls against an agent before it reaches customers, and observability dashboards flag failed calls. Web and mobile SDKs cover in-app voice. Common uses are customer support lines, lead qualification, appointment scheduling and outbound sales calls.

Vapi expects you to write prompts, configure tools and read logs, so non-technical teams usually need a developer or an agency to get an agent into production. Total cost is harder to predict than a flat per-minute rate because it depends on which models you pick. Retell AI offers a similar developer platform with simpler bundled pricing and strong turn-taking, and Bland AI targets enterprises that want a more managed service with its own hosted models. ElevenLabs Agents is the easier choice if you already use ElevenLabs voices and want one vendor. Pick Vapi if you have engineers, expect high call volumes and want freedom to change speech, language and voice providers as models improve. Choose Retell or a no-code tool if your team wants a working receptionist this week without touching an API.