Client Constructor

<ParamField path="provider" type="string" required> This configures which provider to use. The provider is responsible for handling the actual API calls to the LLM service. The provider is a required field.

The configuration modifies the URL request BAML runtime makes.

Provider Name	Docs	Notes
anthropic	Anthropic	Supports /v1/messages endpoint
aws-bedrock	AWS Bedrock	Supports Converse and ConverseStream endpoint
google-ai	Google AI	Supports Google AI's generateContent and streamGenerateContent endpoints
vertex-ai	Vertex AI	Supports Vertex's generateContent and streamGenerateContent endpoints
openai	OpenAI	Supports /chat/completions endpoint
openai-responses	OpenAI Responses API	Supports OpenAI's most advanced /responses endpoint
azure-openai	Azure OpenAI	Supports Azure's /chat/completions endpoint
openai-generic	OpenAI (generic)	Any other provider that supports OpenAI's /chat/completions endpoint

A non-exhaustive list of providers you can use with openai-generic:

Inference Provider	Docs
Azure AI Foundry	Azure AI Foundry
Groq	Groq
Hugging Face	Hugging Face
Keywords AI	Keywords AI
Litellm	Litellm
LM Studio	LM Studio
Ollama	Ollama
OpenRouter	OpenRouter
Vercel AI Gateway	Vercel AI Gateway
TogetherAI	TogetherAI
Unify AI	Unify AI
vLLM	vLLM

We also have some special providers that allow composing clients together:

Provider Name	Docs	Notes
fallback	Fallback	Used to chain models conditional on failures
round-robin	Round Robin	Used to load balance

</ParamField> <ParamField path="options" type="dict[str, Any]" required> These vary per provider. Please see provider specific documentation for more information. Generally they are pass through options to the POST request made to the LLM. </ParamField>