Create assistant

The provider of the assistant’s LLM. Options: openai, anthropic, google, deep-seek, perplexity-ai, vapi, anyscale, together-ai, inflection-ai, groq, xai, deepinfra

The model name (e.g., gpt-4o-mini, deepseek-chat, gemini-1.5-flash, claude-3-haiku-20240307) must match the provider and allowed values in Vapi’s API.

The role of the assistant. Options: assistant, function, user, system, tool

The prompt for the assistant

The name of the assistant

How many seconds of silence to wait before ending the call. Must be between 10 and 3600 seconds

Maximum number of seconds that the call will last. Must be between 10 and 43200 seconds

Enable recording for the call

Determines whether the transcript is stored

The name of the assistant in the transcript

The name of the user in the transcript

Enables filtering of noise and background speech while the user is talking

Background sound in the call. Options: off, office

Settings to configure or disable voicemail detection

Message that the assistant will say if the call is forwarded to voicemail. If unspecified, it will hang up

Message that the assistant will say if it ends the call (max 1000 characters). If unspecified, it will hang up without saying anything

List of phrases that, if spoken by the assistant, will trigger the call to be hung up. Case insensitive

Options: assistant-speaks-first, assistant-speaks-first-with-model-generated-message, assistant-waits-for-user

First message that the assistant will say. If unspecified, assistant will wait for user to speak and use the model to respond

Enable the plan to customize the plan for the summary

The messages used to generate the summary

Unique identifier for the assistant

Organization identifier

When the assistant was created

When the assistant was last updated

Whether an error occurred

Error message if an error occurred

Error status code if an error occurred