Assistant
This is the assistant that will be used for the call. To use an existing assistant, use `assistantId` instead.
If you're unsure why you're getting an invalid assistant, try logging your response and send the JSON blob to POST /assistant which will return the validation errors.
Properties
This determines whether the model says 'mhmm', 'ahem' etc. while user is speaking.
Default false while in beta.
This enables filtering of noise and background speech while the user is talking.
Default false while in beta.
This is the background sound in the call. Default for phone calls is 'office' and default for web calls is 'off'.
These are the messages that will be sent to your Client SDKs. Default is CONVERSATION_UPDATE, FUNCTION_CALL, HANG, MODEL_OUTPUT, SPEECH_UPDATE, STATUS_UPDATE, TRANSCRIPT, TOOL_CALLS, USER_INTERRUPTED, and VOICE_INPUT. You can check the shape of the messages in ClientMessage schema.
This is the message that the assistant will say if it ends the call.
If unspecified, it will hang up without saying anything.
This is the first message that the assistant will say. This can also be a URL to a containerized audio file (mp3, wav, etc.).
If unspecified, assistant will wait for user to speak and use the model to respond once they speak.
This is the mode for the first message. Default is 'assistant-speaks-first'. Use:
assistant.model.messages at call start, call.messages at squad transfer points).
@default 'assistant-speaks-first'
When this is enabled, no logs, recordings, or transcriptions will be stored. At the end of the call, you will still receive an end-of-call-report message to store on your server. Defaults to false.
The minimum number of seconds to wait after transcription (with punctuation) before sending a request to the model. Defaults to 0.1.
@default 0.1
The minimum number of seconds to wait after transcription (without punctuation) before sending a request to the model. Defaults to 1.5.
@default 1.5
This is the maximum number of seconds that the call will last. When the call reaches this duration, it will be ended.
@default 1800 (~30 minutes)
This determines whether the model's output is used in conversation history rather than the transcription of assistant's speech.
Default false while in beta.
The number of words to wait for before interrupting the assistant.
Words like "stop", "actually", "no", etc. will always interrupt immediately regardless of this value.
Words like "okay", "yeah", "right" will never interrupt.
When set to 0, it will rely solely on the VAD (Voice Activity Detector) and will not wait for any transcription. Defaults to this (0).
@default 0
This sets whether the assistant's calls are recorded. Defaults to true.
The minimum number of seconds after user speech to wait before the assistant starts speaking. Defaults to 0.4.
@default 0.4
These are the messages that will be sent to your Server URL. Default is CONVERSATION_UPDATE, END_OF_CALL_REPORT, FUNCTION_CALL, HANG, SPEECH_UPDATE, STATUS_UPDATE, TOOL_CALLS, TRANSFER_DESTINATION_REQUEST, USER_INTERRUPTED. You can check the shape of the messages in ServerMessage schema.
This is the URL Vapi will communicate with via HTTP GET and POST Requests. This is used for retrieving context, function calling, and end-of-call reports.
All requests will be sent with the call object among other things relevant to that message. You can find more details in the Server URL documentation.
This overrides the serverUrl set on the org and the phoneNumber. Order of precedence: tool.server.url > assistant.serverUrl > phoneNumber.serverUrl > org.serverUrl
This is the secret you can set that Vapi will send with every request to your server. Will be sent as a header called x-vapi-secret.
Same precedence logic as serverUrl.
How many seconds of silence to wait before ending the call. Defaults to 30.
This determines whether the video is recorded during the call. Default is false. Only relevant for webCall type.
This is the message that the assistant will say if the call is forwarded to voicemail.
If unspecified, it will hang up.
Functions
This is the plan for analysis of assistant's calls. Stored in call.analysis.
Builder for the Anthropic model.
Builder for the Anyscale model.
This is the plan for artifacts generated during assistant's calls. Stored in call.artifact.
Note: recordingEnabled is currently at the root level. It will be moved to artifactPlan in the future, but will remain backwards compatible.
These are the overrides for the assistant's settings and template variables.
Builder for the Azure voice.
Builder for the Cartesia voice.
Builder for the CustomLLM model.
Builder for the Deepgram transcriber.
Builder for the Deepgram voice.
Builder for the DeepInfra model.
Builder for the ElevenLabs voice.
Builder for the Gladia transcriber.
Builder for the Neets voice.
Builder for the OpenAI model.
Builder for the OpenAI voice.
Builder for the OpenRouter model.
Builder for the PerplexityAI model.
Builder for the PlayHT voice.
Builder for the RimeAI voice.
Builder for the Talkscriber transcriber.
Builder for the TogetherAI model.
These are the settings to configure or disable voicemail detection. Alternatively, voicemail detection can be configured using the model.tools=VoicemailTool. This uses Twilio's built-in detection while the VoicemailTool relies on the model to detect if a voicemail was reached. You can use neither of them, one of them, or both of them. By default, Twilio built-in detection is enabled while VoicemailTool is not.