ResourcesIntegrationsLarge language models

Anyscale

Use an OpenAI-compatible Anyscale endpoint as the language model for your Vapi voice agent.

Anyscale is a large language model provider whose OpenAI-compatible endpoints serve open models. Connect Anyscale to use one as the LLM on a Vapi assistant. This page covers what Anyscale is and how to use it.

What is Anyscale

Anyscale serves open large language models through an OpenAI-compatible API. Anyscale has no preset model list in Vapi. You enter the name of the model your Anyscale endpoint provides.

You can use Anyscale through Vapi’s default integration, or use your own Anyscale account by connecting it in Integrations.

Language model

Set model.provider to anyscale and model.model to a model your Anyscale endpoint serves.

$curl -X PATCH "https://api.vapi.ai/assistant/ASSISTANT_ID" \
>  -H "Authorization: Bearer $VAPI_API_KEY" \
>  -H "Content-Type: application/json" \
>  -d '{
>    "model": { "provider": "anyscale", "model": "YOUR_MODEL" }
>  }'

For additional API configuration options, review the AnyscaleModel fields in the Create Assistant API reference.

Model providers

Compare every LLM provider Vapi supports.