bedrock.AsyncClient

Async_Client utilities for aws bedrock integration utilities.

View module source on GitHub

MaximBedrockAsyncClient

class MaximBedrockAsyncClient()

Maxim Bedrock async client wrapper.

This class provides a wrapper around the Bedrock async client to integrate with Maxim’s logging and monitoring capabilities. It allows tracking and logging of Bedrock API interactions through the Maxim platform.

__init__

def __init__(logger: Logger, client: BaseClient)

Initialize the Maxim Bedrock async client.

Arguments:

Name	Type	Description
logger	Logger	The Maxim logger instance for tracking interactions.
client	BaseClient	The Bedrock async client instance to wrap.

converse_stream

async def converse_stream(
        messages: Iterable[BedrockMessageParam],
        system: Optional[Union[str, List[Dict[str, str]]]] = None,
        *,
        model: str,
        modelId: str,
        max_tokens: Optional[int] = None,
        trace_id: Optional[str] = None,
        generation_name: Optional[str] = None,
        inferenceConfig: Optional[Dict[str, Any]] = None,
        toolConfig: Optional[Dict[str, Any]] = None,
        guardrailConfig: Optional[Dict[str, Any]] = None,
        performanceConfig: Optional[Dict[str, Any]] = None,
        additionalModelRequestFields: Optional[Dict[str, Any]] = None,
        **kwargs: Any) -> AsyncGenerator[BedrockStreamEvent, None]

Converse with the Bedrock client.

Arguments:

Name	Type	Description
messages	Iterable[BedrockMessageParam]	The messages to send to the Bedrock client.
system	Optional[Union[str, List[Dict[str, str]]]]	The system message to send to the Bedrock client.
model	str	The model to use for the Bedrock client.
modelId	str	The model ID to use for the Bedrock client.
max_tokens	Optional[int]	The maximum number of tokens to generate.
trace_id	Optional[str]	The trace ID to use for the Bedrock client.
generation_name	Optional[str]	The name of the generation to use for the Bedrock client.
inferenceConfig	Optional[Dict[str, Any]]	The inference configuration to use for the Bedrock client.
toolConfig	Optional[Dict[str, Any]]	The tool configuration to use for the Bedrock client.
guardrailConfig	Optional[Dict[str, Any]]	The guardrail configuration to use for the Bedrock client.
performanceConfig	Optional[Dict[str, Any]]	The performance configuration to use for the Bedrock client.
additionalModelRequestFields	Optional[Dict[str, Any]]	The additional model request fields to use for the Bedrock client.

Returns:

AsyncGenerator[BedrockStreamEvent, None]: An asynchronous generator over the Bedrock stream events.

converse

async def converse(messages: Iterable[BedrockMessageParam],
                   system: Optional[Union[str, List[Dict[str, str]]]] = None,
                   *,
                   model: str,
                   modelId: str,
                   max_tokens: Optional[int] = None,
                   trace_id: Optional[str] = None,
                   generation_name: Optional[str] = None,
                   inferenceConfig: Optional[Dict[str, Any]] = None,
                   toolConfig: Optional[Dict[str, Any]] = None,
                   guardrailConfig: Optional[Dict[str, Any]] = None,
                   performanceConfig: Optional[Dict[str, Any]] = None,
                   additionalModelRequestFields: Optional[Dict[str,
                                                               Any]] = None,
                   **kwargs: Any) -> BedrockMessage

Converse with the Bedrock client.

Arguments:

Name	Type	Description
messages	Iterable[BedrockMessageParam]	The messages to send to the Bedrock client.
system	Optional[Union[str, List[Dict[str, str]]]]	The system message to send to the Bedrock client.
model	str	The model to use for the Bedrock client.
modelId	str	The model ID to use for the Bedrock client.
max_tokens	Optional[int]	The maximum number of tokens to generate.
trace_id	Optional[str]	The trace ID to use for the Bedrock client.
generation_name	Optional[str]	The name of the generation to use for the Bedrock client.
inferenceConfig	Optional[Dict[str, Any]]	The inference configuration to use for the Bedrock client.
toolConfig	Optional[Dict[str, Any]]	The tool configuration to use for the Bedrock client.
guardrailConfig	Optional[Dict[str, Any]]	The guardrail configuration to use for the Bedrock client.
performanceConfig	Optional[Dict[str, Any]]	The performance configuration to use for the Bedrock client.
additionalModelRequestFields	Optional[Dict[str, Any]]	The additional model request fields to use for the Bedrock client.

Returns:

Name	Description
BedrockMessage	The response from the Bedrock client.