Model providers

Look up the LLM providers Vapi supports, their IDs, and their compliance coverage.

This page lists the large language model (LLM) providers Vapi supports, with each provider’s ID, compliance coverage, and dedicated documentation.

The model is the LLM that generates your voice agent’s responses. You select one by setting model.provider and model.model on an assistant. See Create Assistant in the API reference.

Quick reference

Every column below reflects Vapi’s coverage for that provider. The provider string is the exact value you pass in model.provider. Custom LLM is N/A because it runs on your own server, so whether it meets HIPAA, GDPR, or PCI is up to you to verify.

ProviderProvider stringHIPAAGDPR (EU region)PCI
AnthropicanthropicYesNoNo
Anthropic (Bedrock)anthropic-bedrockYesYesYes
AnyscaleanyscaleNoNoNo
Azure OpenAIazure-openaiYesYesYes
CerebrascerebrasNoNoNo
Custom LLMcustom-llmN/AN/AN/A
DeepInfradeepinfraNoNoNo
DeepSeekdeep-seekNoNoNo
GooglegoogleYesYesYes
GroqgroqNoNoNo
Inflection AIinflection-aiNoNoNo
MiniMaxminimaxNoNoNo
MistralmistralNoNoNo
OpenAIopenaiYesYesYes
OpenRouteropenrouterNoNoNo
Perplexityperplexity-aiNoNoNo
Together AItogether-aiYesNoNo
xAIxaiYesYesNo

The Assistant → Model dropdown in the Vapi Dashboard is the most current list of selectable models.