Skip to navigation
ResourcesIntegrationsLarge language models

Groq

Use Groq’s low-latency models as the language model for your Vapi voice agent

Groq provides low-latency inference for supported open and partner models. In Vapi, Groq acts as the language model that generates your assistant’s responses.

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

Language model

Set model.provider to groq and model.model to one of the supported models below.

curl -X PATCH "https://api.vapi.ai/assistant/ASSISTANT_ID" \
  -H "Authorization: Bearer $VAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": { "provider": "groq", "model": "openai/gpt-oss-120b" }
  }'

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

These are the Groq models currently selectable in Vapi.

ModelModel ID
GPT-OSS 20Bopenai/gpt-oss-20b
GPT-OSS 120Bopenai/gpt-oss-120b
Groq models

Check current upstream availability and model lifecycle information.

Language models

Compare every language model provider Vapi supports.