converse_stream async ¶

@asynccontextmanager
async def converse_stream(
    self,
    model_id: "capo_bedrock_runtime.types.conversational_model_id.ConversationalModelId",
    *,
    config_overrides: Optional[AsyncBedrockRuntimeClientConfig] = None,
    messages: Optional["capo_bedrock_runtime.types.messages.Messages"] = None,
    system: Optional[
        "capo_bedrock_runtime.types.system_content_blocks.SystemContentBlocks"
    ] = None,
    inference_config: Optional[
        "capo_bedrock_runtime.types.inference_configuration.InferenceConfiguration"
    ] = None,
    tool_config: Optional[
        "capo_bedrock_runtime.types.tool_configuration.ToolConfiguration"
    ] = None,
    guardrail_config: Optional[
        "capo_bedrock_runtime.types.guardrail_stream_configuration.GuardrailStreamConfiguration"
    ] = None,
    additional_model_request_fields: Optional[object] = None,
    prompt_variables: Optional[
        "capo_bedrock_runtime.types.prompt_variable_map.PromptVariableMap"
    ] = None,
    additional_model_response_field_paths: Optional[
        "capo_bedrock_runtime.types.additional_model_response_field_paths.AdditionalModelResponseFieldPaths"
    ] = None,
    request_metadata: Optional[
        "capo_bedrock_runtime.types.request_metadata.RequestMetadata"
    ] = None,
    performance_config: Optional[
        "capo_bedrock_runtime.types.performance_configuration.PerformanceConfiguration"
    ] = None,
    service_tier: Optional[
        "capo_bedrock_runtime.types.service_tier.ServiceTier"
    ] = None,
    output_config: Optional[
        "capo_bedrock_runtime.types.output_config.OutputConfig"
    ] = None,
) -> "AsyncGenerator[capo_bedrock_runtime.types.converse_stream_response.ConverseStreamResponse]":
    """<p>Sends messages to the specified Amazon Bedrock model and returns the response in a stream. <code>ConverseStream</code> provides a consistent API that works with all Amazon Bedrock models that support messages. This allows you to write code once and use it with different models. Should a model have unique inference parameters, you can also pass those unique parameters to the model. </p> <p>To find out if a model supports streaming, call <a href="https://docs.aws.amazon.com/bedrock/latest/APIReference/API_GetFoundationModel.html">GetFoundationModel</a> and check the <code>responseStreamingSupported</code> field in the response.</p> <note> <p>The CLI doesn't support streaming operations in Amazon Bedrock, including <code>ConverseStream</code>.</p> </note> <p>Amazon Bedrock doesn't store any text, images, or documents that you provide as content. The data is only used to generate the response.</p> <p>You can submit a prompt by including it in the <code>messages</code> field, specifying the <code>modelId</code> of a foundation model or inference profile to run inference on it, and including any other fields that are relevant to your use case.</p> <p>You can also submit a prompt from Prompt management by specifying the ARN of the prompt version and including a map of variables to values in the <code>promptVariables</code> field. You can append more messages to the prompt by using the <code>messages</code> field. If you use a prompt from Prompt management, you can't include the following fields in the request: <code>additionalModelRequestFields</code>, <code>inferenceConfig</code>, <code>system</code>, or <code>toolConfig</code>. Instead, these fields must be defined through Prompt management. For more information, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-management-use.html">Use a prompt from Prompt management</a>.</p> <p>For information about the Converse API, see <i>Use the Converse API</i> in the <i>Amazon Bedrock User Guide</i>. To use a guardrail, see <i>Use a guardrail with the Converse API</i> in the <i>Amazon Bedrock User Guide</i>. To use a tool with a model, see <i>Tool use (Function calling)</i> in the <i>Amazon Bedrock User Guide</i> </p> <p>For example code, see <i>Conversation streaming example</i> in the <i>Amazon Bedrock User Guide</i>. </p> <p>This operation requires permission for the <code>bedrock:InvokeModelWithResponseStream</code> action.</p> <important> <p>To deny all inference access to resources that you specify in the modelId field, you need to deny access to the <code>bedrock:InvokeModel</code> and <code>bedrock:InvokeModelWithResponseStream</code> actions. Doing this also denies access to the resource through the base inference actions (<a href="https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModel.html">InvokeModel</a> and <a href="https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_InvokeModelWithResponseStream.html">InvokeModelWithResponseStream</a>). For more information see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/security_iam_id-based-policy-examples.html#security_iam_id-based-policy-examples-deny-inference">Deny access for inference on specific models</a>. </p> </important> <p>For troubleshooting some of the common errors you might encounter when using the <code>ConverseStream</code> API, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html">Troubleshooting Amazon Bedrock API Error Codes</a> in the Amazon Bedrock User Guide</p>

    Args:
        model_id: <p>Specifies the model or throughput with which to run inference, or the prompt resource to use in inference. The value depends on the resource that you use:</p> <ul> <li> <p>If you use a base model, specify the model ID or its ARN. For a list of model IDs for base models, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids.html#model-ids-arns">Amazon Bedrock base model IDs (on-demand throughput)</a> in the Amazon Bedrock User Guide.</p> </li> <li> <p>If you use an inference profile, specify the inference profile ID or its ARN. For a list of inference profile IDs, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference-support.html">Supported Regions and models for cross-region inference</a> in the Amazon Bedrock User Guide.</p> </li> <li> <p>If you use a provisioned model, specify the ARN of the Provisioned Throughput. For more information, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/prov-thru-use.html">Run inference using a Provisioned Throughput</a> in the Amazon Bedrock User Guide.</p> </li> <li> <p>If you use a custom model, first purchase Provisioned Throughput for it. Then specify the ARN of the resulting provisioned model. For more information, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-customization-use.html">Use a custom model in Amazon Bedrock</a> in the Amazon Bedrock User Guide.</p> </li> <li> <p>To include a prompt that was defined in <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-management.html">Prompt management</a>, specify the ARN of the prompt version to use.</p> </li> </ul> <p>The Converse API doesn't support <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-customization-import-model.html">imported models</a>.</p>
        messages: <p>The messages that you want to send to the model.</p>
        system: <p>A prompt that provides instructions or context to the model about the task it should perform, or the persona it should adopt during the conversation.</p>
        inference_config: <p>Inference parameters to pass to the model. <code>Converse</code> and <code>ConverseStream</code> support a base set of inference parameters. If you need to pass additional parameters that the model supports, use the <code>additionalModelRequestFields</code> request field.</p>
        tool_config: <p>Configuration information for the tools that the model can use when generating a response.</p> <p>For information about models that support streaming tool use, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/conversation-inference.html#conversation-inference-supported-models-features">Supported models and model features</a>.</p>
        guardrail_config: <p>Configuration information for a guardrail that you want to use in the request. If you include <code>guardContent</code> blocks in the <code>content</code> field in the <code>messages</code> field, the guardrail operates only on those messages. If you include no <code>guardContent</code> blocks, the guardrail operates on all messages in the request body and in any included prompt resource.</p>
        additional_model_request_fields: <p>Additional inference parameters that the model supports, beyond the base set of inference parameters that <code>Converse</code> and <code>ConverseStream</code> support in the <code>inferenceConfig</code> field. For more information, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/model-parameters.html">Model parameters</a>.</p>
        prompt_variables: <p>Contains a map of variables in a prompt from Prompt management to objects containing the values to fill in for them when running model invocation. This field is ignored if you don't specify a prompt resource in the <code>modelId</code> field.</p>
        additional_model_response_field_paths: <p>Additional model parameters field paths to return in the response. <code>Converse</code> and <code>ConverseStream</code> return the requested fields as a JSON Pointer object in the <code>additionalModelResponseFields</code> field. The following is example JSON for <code>additionalModelResponseFieldPaths</code>.</p> <p> <code>[ "/stop_sequence" ]</code> </p> <p>For information about the JSON Pointer syntax, see the <a href="https://datatracker.ietf.org/doc/html/rfc6901">Internet Engineering Task Force (IETF)</a> documentation.</p> <p> <code>Converse</code> and <code>ConverseStream</code> reject an empty JSON Pointer or incorrectly structured JSON Pointer with a <code>400</code> error code. if the JSON Pointer is valid, but the requested field is not in the model response, it is ignored by <code>Converse</code>.</p>
        request_metadata: <p>Key-value pairs that you can use to filter invocation logs.</p>
        performance_config: <p>Model performance settings for the request.</p>
        service_tier: <p>Specifies the processing tier configuration used for serving the request.</p>
        output_config: <p>Output configuration for a model response.</p>

    Raises:
        capo_bedrock_runtime.errors.access_denied_exception.AccessDeniedException: <p>The request is denied because you do not have sufficient permissions to perform the requested action. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-access-denied">AccessDeniedException</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.internal_server_exception.InternalServerException: <p>An internal server error occurred. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-internal-failure">InternalFailure</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.model_error_exception.ModelErrorException: <p>The request failed due to an error while processing the model.</p>
        capo_bedrock_runtime.errors.model_not_ready_exception.ModelNotReadyException: <p>The model specified in the request is not ready to serve inference requests. The AWS SDK will automatically retry the operation up to 5 times. For information about configuring automatic retries, see <a href="https://docs.aws.amazon.com/sdkref/latest/guide/feature-retry-behavior.html">Retry behavior</a> in the <i>AWS SDKs and Tools</i> reference guide.</p>
        capo_bedrock_runtime.errors.model_timeout_exception.ModelTimeoutException: <p>The request took too long to process. Processing time exceeded the model timeout length.</p>
        capo_bedrock_runtime.errors.resource_not_found_exception.ResourceNotFoundException: <p>The specified resource ARN was not found. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-resource-not-found">ResourceNotFound</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.service_unavailable_exception.ServiceUnavailableException: <p>The service isn't currently available. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-service-unavailable">ServiceUnavailable</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.throttling_exception.ThrottlingException: <p>Your request was denied due to exceeding the account quotas for <i>Amazon Bedrock</i>. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-throttling-exception">ThrottlingException</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.validation_exception.ValidationException: <p>The input fails to satisfy the constraints specified by <i>Amazon Bedrock</i>. For troubleshooting this error, see <a href="https://docs.aws.amazon.com/bedrock/latest/userguide/troubleshooting-api-error-codes.html#ts-validation-error">ValidationError</a> in the Amazon Bedrock User Guide</p>
        capo_bedrock_runtime.errors.UnknownServiceError: The service returned an error code this client does not model.
    """

    async def _handler(
        req: "AsyncOperationRequest[capo_bedrock_runtime.types.converse_stream_request.ConverseStreamRequest]",
    ) -> AsyncOperationResponse[
        "capo_bedrock_runtime.types.converse_stream_response.ConverseStreamResponse"
    ]:
        import capo_bedrock_runtime._operations.amazon_bedrock_frontend_service.converse_stream

        (
            output,
            http_response,
        ) = await capo_bedrock_runtime._operations.amazon_bedrock_frontend_service.converse_stream.async_converse_stream(
            req.options, req.input
        )
        return AsyncOperationResponse(output=output, response=http_response)

    def _redact_input(
        input_: "capo_bedrock_runtime.types.converse_stream_request.ConverseStreamRequest",
    ) -> dict[str, Any]:
        import capo_bedrock_runtime.types.converse_stream_request

        return capo_bedrock_runtime.types.converse_stream_request.redact_sensitive(
            input_
        )

    def _redact_output(
        output: "capo_bedrock_runtime.types.converse_stream_response.ConverseStreamResponse",
    ) -> dict[str, Any]:
        import capo_bedrock_runtime.types.converse_stream_response

        return capo_bedrock_runtime.types.converse_stream_response.redact_sensitive(
            output
        )

    interceptors_, options_ = self.operation_options(
        config_overrides,
        OperationLogging(
            "ConverseStream",
            redact_input=_redact_input,
            redact_output=_redact_output,
        ),
    )
    input_: capo_bedrock_runtime.types.converse_stream_request.ConverseStreamRequest = {
        "model_id": model_id
    }
    if messages is not None:
        input_["messages"] = messages
    if system is not None:
        input_["system"] = system
    if inference_config is not None:
        input_["inference_config"] = inference_config
    if tool_config is not None:
        input_["tool_config"] = tool_config
    if guardrail_config is not None:
        input_["guardrail_config"] = guardrail_config
    if additional_model_request_fields is not None:
        input_["additional_model_request_fields"] = additional_model_request_fields
    if prompt_variables is not None:
        input_["prompt_variables"] = prompt_variables
    if additional_model_response_field_paths is not None:
        input_["additional_model_response_field_paths"] = (
            additional_model_response_field_paths
        )
    if request_metadata is not None:
        input_["request_metadata"] = request_metadata
    if performance_config is not None:
        input_["performance_config"] = performance_config
    if service_tier is not None:
        input_["service_tier"] = service_tier
    if output_config is not None:
        input_["output_config"] = output_config

    response = await aexecute_pipeline(
        AsyncOperationRequest(input=input_, options=options_),
        handler=_handler,
        interceptors=list(interceptors_),
    )
    try:
        yield response.output
    finally:
        await response.response.aclose()