capo-sagemaker-inference-router-runtime: typed, async Python SDK for AWS Sagemaker Inference Router Runtime¶
capo-sagemaker-inference-router-runtime is a typed Python SDK for AWS Sagemaker Inference Router Runtime. It has an async client, AsyncSageMakerInferenceRouterRuntimeClient, for asyncio and trio, and a sync client, SageMakerInferenceRouterRuntimeClient, with the same methods. Requests and responses are typed with TypedDicts, so your editor and type checker know every field.
It is part of capo, a community-driven alternative to boto3 and aioboto3, built on its own HTTP stack instead of botocore and generated from the official AWS Smithy models. Documentation: https://capo-sdk.dev/services/sagemaker-inference-router-runtime/
Installation¶
Usage¶
from capo_sagemaker_inference_router_runtime import AsyncSageMakerInferenceRouterRuntimeClient
async def main():
async with AsyncSageMakerInferenceRouterRuntimeClient() as sage_maker_inference_router_runtime:
# Example: call the invoke_inference_router operation
response = await sage_maker_inference_router_runtime.invoke_inference_router()
print(response["body"])
Error Handling¶
The SDK raises exceptions for errors returned by the API. Catch them to handle failures gracefully.
from capo_sagemaker_inference_router_runtime import AsyncSageMakerInferenceRouterRuntimeClient
from capo_sagemaker_inference_router_runtime.error import AccessDeniedException
async def main():
async with AsyncSageMakerInferenceRouterRuntimeClient() as sage_maker_inference_router_runtime:
try:
await sage_maker_inference_router_runtime.invoke_inference_router()
except AccessDeniedException as e:
print(f"Error: {e}")
print(e.data) # additional error data
Retrying¶
The SDK retries failed operations automatically. Retry behaviour follows the Smithy specification: errors are retried based on their is_retryable and is_throttling_error attributes. Throttling errors use a longer base delay. Network-level failures (connection errors and timeouts) are also retried. Non-retryable errors, such as client errors without the @retryable trait, are raised immediately without further attempts.
The number of attempts defaults to 3 and can be changed at the client level via retry_max_attempts, or per call via config_overrides.
retry_mode="adaptive" (or AWS_RETRY_MODE=adaptive, or retry_mode = adaptive in the AWS profile) also slows the whole client down once the service starts throttling it, then speeds back up as requests succeed. The default is standard.
from capo_sagemaker_inference_router_runtime import AsyncSageMakerInferenceRouterRuntimeClient
async def main():
async with AsyncSageMakerInferenceRouterRuntimeClient() as sage_maker_inference_router_runtime:
# Default: 3 attempts for every operation
response = await sage_maker_inference_router_runtime.invoke_inference_router()
# Override per operation
response = await sage_maker_inference_router_runtime.invoke_inference_router(config_overrides={"retry_max_attempts": 5})
# Disable retries for this call
response = await sage_maker_inference_router_runtime.invoke_inference_router(config_overrides={"retry_max_attempts": 1})