Skip to main content

​
SambaModel

class SambaModel(BaseModelBackend):
SambaNova service interface. Parameters:
  • model_type (Union[ModelType, str]): Model for which a SambaNova backend is created. Supported models via SambaNova Cloud: https://community.sambanova.ai/t/supported-models/193. Supported models via SambaVerse API is listed in https://sambaverse.sambanova.ai/models.
  • model_config_dict (Optional[Dict[str, Any]], optional): A dictionary that will be fed into:obj:openai.ChatCompletion.create(). If :obj:None, :obj:SambaCloudAPIConfig().as_dict() will be used. (default: :obj:None)
  • api_key (Optional[str], optional): The API key for authenticating with the SambaNova service. (default: :obj:None)
  • url (Optional[str], optional): The url to the SambaNova service. Current support SambaVerse API: :obj:"https://sambaverse.sambanova.ai/api/predict" and SambaNova Cloud: :obj:"https://api.sambanova.ai/v1" (default: :obj:https://api. sambanova.ai/v1)
  • token_counter (Optional[BaseTokenCounter], optional): Token counter to use for the model. If not provided, :obj:OpenAITokenCounter( ModelType.GPT_4O_MINI) will be used.
  • timeout (Optional[float], optional): The timeout value in seconds for API calls. If not provided, will fall back to the MODEL_TIMEOUT environment variable or default to 180 seconds. (default: :obj:None)
  • max_retries (int, optional): Maximum number of retries for API calls. (default: :obj:3)
  • client (Optional[Any], optional): A custom synchronous OpenAI-compatible client instance. If provided, this client will be used instead of creating a new one. Only applicable when using SambaNova Cloud API. (default: :obj:None)
  • async_client (Optional[Any], optional): A custom asynchronous OpenAI-compatible client instance. If provided, this client will be used instead of creating a new one. Only applicable when using SambaNova Cloud API. (default: :obj:None) **kwargs (Any): Additional arguments to pass to the client initialization. Ignored if custom clients are provided.

​
init

def __init__(
    self,
    model_type: Union[ModelType, str],
    model_config_dict: Optional[Dict[str, Any]] = None,
    api_key: Optional[str] = None,
    url: Optional[str] = None,
    token_counter: Optional[BaseTokenCounter] = None,
    timeout: Optional[float] = None,
    max_retries: int = 3,
    client: Optional[Any] = None,
    async_client: Optional[Any] = None,
    **kwargs: Any
):

​
token_counter

def token_counter(self):
Returns: BaseTokenCounter: The token counter following the model’s tokenization style.

​
_run

def _run(
    self,
    messages: List[OpenAIMessage],
    response_format: Optional[Type[BaseModel]] = None,
    tools: Optional[List[Dict[str, Any]]] = None
):
Runs SambaNova’s service. Parameters:
  • messages (List[OpenAIMessage]): Message list with the chat history in OpenAI API format.
Returns: Union[ChatCompletion, Stream[ChatCompletionChunk]]: ChatCompletion in the non-stream mode, or Stream[ChatCompletionChunk] in the stream mode.

​
_run_streaming

def _run_streaming(self, messages: List[OpenAIMessage]):
Handles streaming inference with SambaNova’s API. Parameters:
  • messages (List[OpenAIMessage]): A list of messages representing the chat history in OpenAI API format.
Returns: Stream[ChatCompletionChunk]: A generator yielding ChatCompletionChunk objects as they are received from the API.

​
_run_non_streaming

def _run_non_streaming(self, messages: List[OpenAIMessage]):
Handles non-streaming inference with SambaNova’s API. Parameters:
  • messages (List[OpenAIMessage]): A list of messages representing the message in OpenAI API format.
Returns: ChatCompletion: A ChatCompletion object containing the complete response from the API.

​
_sambaverse_to_openai_response

def _sambaverse_to_openai_response(self, samba_response: Dict[str, Any]):
Converts SambaVerse API response into an OpenAI-compatible response. Parameters:
  • samba_response (Dict[str, Any]): A dictionary representing responses from the SambaVerse API.
Returns: ChatCompletion: A ChatCompletion object constructed from the aggregated response data.

​
stream

def stream(self):
Returns: bool: Whether the model is in stream mode.