SageMaker Asynchronous Inference Max Processing Time

Hello,

I can see from the AWS documentation that it mentions “long processing times (up to one hour)” and the second document you referred suggests that max processing time is upto 15 minutes.

To remove the ambiguity, the default processing time for SageMaker Asynchronous Inference is 15 minutes. Total processing time can be extended up to 3600 seconds (60 minutes). [1]


To do so, “InvocationTimeoutSeconds=3600" parameter must be used when invoking the asynchronous endpoint. If you don't specify this field in your request, by default the request times out at 15 minutes. [2]

					response = sagemaker_runtime.invoke_endpoint_async(
								EndpointName=endpoint_name,
								InputLocation=input_location,
								InvocationTimeoutSeconds=3600)

References:

[1] InvokeEndpointAsync - https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_runtime_InvokeEndpointAsync.html

[2] Invoke an Asynchronous Endpoint - https://docs.aws.amazon.com/sagemaker/latest/dg/async-inference-invoke-endpoint.html