| Parameter | Type | Description | Default | 
|---|---|---|---|
max_completion_tokens | Optional[int] | Maximum tokens in completion | None | 
repetition_penalty | Optional[float] | Penalty for token repetition | None | 
temperature | Optional[float] | Sampling temperature | None | 
top_p | Optional[float] | Nucleus sampling parameter | None | 
top_k | Optional[int] | Top-k sampling parameter | None | 
extra_headers | Optional[Any] | Additional HTTP headers to include in the API request | None | 
extra_query | Optional[Any] | Additional query parameters to include in the API request | None | 
extra_body | Optional[Any] | Additional body parameters to include in the API request | None | 
request_params | Optional[Dict[str, Any]> | Custom request parameters dictionary, merged into the API request | None | 
api_key | Optional[str] | Llama API key (overrides LLAMA_API_KEY environment variable) | None | 
base_url | Optional[Union[str, httpx.URL]] | Base URL for the Llama API | None | 
timeout | Optional[float] | Timeout for API requests (seconds) | None | 
max_retries | Optional[int] | Maximum number of retries for API calls | None | 
default_headers | Optional[Any] | Default HTTP headers for the API client | None | 
default_query | Optional[Any] | Default query parameters for the API client | None | 
http_client | Optional[httpx.Client] | Custom synchronous HTTP client instance | None | 
client_params | Optional[Dict[str, Any]> | Additional parameters for the HTTP client constructor | None |