Create a call with inline configuration
Run a call without a saved agent — the whole configuration travels in the request. Nothing is stored and nothing is reused.
Use it for per-user personas, generated agents, or testing a prompt without leaving a record behind. If the same configuration is used repeatedly, create a real agent instead: you gain version history, tool assignment, a phone number, corpora, and the dashboard.
Authorizations
API key for authentication
Body
The entire prompt. There is no basePrompt/context/ customInstructions layering here — that belongs to saved agents.
Voice name or ID. See GET /voices.
"Mark"
Language code. See GET /languages.
"en"
gemma, llama, glm 0 <= x <= 1Appended to the prompt as the opening line when the agent
speaks first. Ignored when firstSpeaker is user.
agent, user Applies to the agent's first message only.
Hard cap in seconds.
60 <= x <= 7200Seconds of silence before the agent reacts.
10 <= x <= 300hang_up, prompt Which transport the returned websocketUrl is prepared for.
twilio / telnyx / plivo — carrier media streams, G.711 audio.
websocket — your own client; you choose the sample rates and codec.
webrtc — browser transport, normally reached through the SDK.
Any other value returns 400.
twilio, telnyx, plivo, websocket, webrtc Audio parameters. Only read when connectionType is websocket —
telephony transports force G.711 at 8000 Hz regardless of what is sent
here.