Vapi API
Vapi grades A+: self-serve pay-as-you-go signup with a $10 trial credit, published $0.05 per minute platform pricing, public docs, client and server SDKs, a CLI, and an MCP server. Enterprise adds volume pricing and SLAs rather than gating access.
Frequently asked questions
Vapi scores A+ on the API Report Card. Vapi grades A+: self-serve pay-as-you-go signup with a $10 trial credit, published $0.05 per minute platform pricing, public docs, client and server SDKs, a CLI, and an MCP server. Enterprise adds volume pricing and SLAs rather than gating access.
Vapi has a workable official integration path. Most engineering teams can build against it directly. Open API: self-serve, documented, with SDKs
Vapi is a voice AI agent platform that lets developers build, deploy, and orchestrate phone and in-app voice agents through an API.
Software and data tools. Developers create assistants through the API or dashboard, attach tools that call their own or third-party endpoints mid-conversation, buy or connect phone numbers, and receive server event callbacks during calls.
Vapi is one of the best-known names in voice agent infrastructure. It is a Y Combinator alum that raised a $50M Series B in May 2026 led by Peak XV Partners at a $500M valuation, with M12, Kleiner Perkins, and Bessemer participating, for roughly $72M raised in total.
Vapi holds assistant configurations, call logs, recordings, and transcripts for its customers' voice agents. It is developer infrastructure rather than a vertical system of record; the business data agents act on lives in the systems the agents call into.
A product of the 2023 to 2024 voice AI wave, built API-first from the start: provider-agnostic model stack, structured outputs, tool calling, a CLI, an MCP server, and pay-as-you-go metering.
Third-party cost analyses note the $0.05 per minute headline covers orchestration only; realistic all-in cost lands near $0.15 to $0.40 per minute once STT, LLM, TTS, and telephony are added. Full sourced list under Sources below.
Common alternatives include Retell AI, Bland AI, ElevenLabs, LiveKit, Pipecat. Graded alternatives appear under "More from the report card" below.
Grades measure one thing: can a customer's engineering team get their own data out programmatically? We check six things (whether a real API exists, how access is gated, data coverage, auth quality, docs and developer experience, and stability) and roll them into a letter grade. Grades get re-verified, and they only move on evidence.