Skip to main content

strip_trailing_whitespace_from_messages

Strip trailing whitespace from all message contents in a list of messages. This is necessary because the Anthropic API doesn’t allow trailing whitespace in message content. Parameters:
  • messages (List[OpenAIMessage]): List of messages to process
Returns: List[OpenAIMessage]: The processed messages with trailing whitespace removed

AnthropicModel

Anthropic API in a unified BaseModelBackend interface. Parameters:
  • model_type (Union[ModelType, str]): Model for which a backend is created, one of CLAUDE_* series.
  • model_config_dict (Optional[Dict[str, Any]], optional): A dictionary that will be fed into Anthropic API. If :obj:None, :obj:AnthropicConfig().as_dict() will be used. (default: :obj:None)
  • api_key (Optional[str], optional): The API key for authenticating with the Anthropic service. (default: :obj:None)
  • url (Optional[str], optional): The url to the Anthropic service. (default: :obj:None)
  • token_counter (Optional[BaseTokenCounter], optional): Token counter to use for the model. If not provided, :obj:AnthropicTokenCounter will be used. (default: :obj:None)
  • 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 Anthropic client instance. If provided, this client will be used instead of creating a new one. (default: :obj:None)
  • async_client (Optional[Any], optional): A custom asynchronous Anthropic client instance. If provided, this client will be used instead of creating a new one. (default: :obj:None)
  • use_beta_for_structured_outputs (bool, optional): Whether to use the beta API for structured outputs. (default: :obj:False) **kwargs (Any): Additional arguments to pass to the client initialization.

init

token_counter

Returns: AnthropicTokenCounter: The token counter following the model’s tokenization style.

_convert_openai_to_anthropic_messages

Convert OpenAI format messages to Anthropic format. Parameters:
  • messages (List[OpenAIMessage]): Messages in OpenAI format.
Returns: tuple[Optional[str], List[Dict[str, Any]]]: A tuple containing the system message (if any) and the list of messages in Anthropic format.

_convert_anthropic_to_openai_response

Convert Anthropic API response to OpenAI ChatCompletion format. Parameters:
  • response: The response object from Anthropic API.
  • model (str): The model name.
Returns: ChatCompletion: Response in OpenAI format.

_convert_anthropic_stream_to_openai_chunk

Convert Anthropic streaming chunk to OpenAI ChatCompletionChunk. Parameters:
  • chunk: The streaming chunk from Anthropic API.
  • model (str): The model name.
  • tool_call_index (Dict[str, int]): A mutable dict tracking tool call indices by their IDs, used to maintain consistent indexing across streaming chunks.
Returns: ChatCompletionChunk: Chunk in OpenAI format.

_convert_openai_tools_to_anthropic

Convert OpenAI tools format to Anthropic tools format. Parameters:
  • tools (Optional[List[Dict[str, Any]]]): Tools in OpenAI format.
Returns: Optional[List[Dict[str, Any]]]: Tools in Anthropic format.

_run

Runs inference of Anthropic chat completion. Parameters:
  • messages (List[OpenAIMessage]): Message list with the chat history in OpenAI API format.
  • response_format (Optional[Type[BaseModel]]): The format of the response. (Not supported by Anthropic API directly)
  • tools (Optional[List[Dict[str, Any]]]): The schema of the tools to use for the request.
Returns: Union[ChatCompletion, Stream[ChatCompletionChunk]]: ChatCompletion in the non-stream mode, or Stream[ChatCompletionChunk] in the stream mode.

_wrap_anthropic_stream

Wrap Anthropic streaming response to OpenAI Stream format. Parameters:
  • stream: The streaming response from Anthropic API.
  • model (str): The model name.
Returns: Stream[ChatCompletionChunk]: Stream in OpenAI format.

_wrap_anthropic_async_stream

Wrap Anthropic async streaming response to OpenAI AsyncStream. Parameters:
  • stream: The async streaming response from Anthropic API.
  • model (str): The model name.
Returns: AsyncStream[ChatCompletionChunk]: AsyncStream in OpenAI format.

stream

Returns: bool: Whether the model is in stream mode.