× We want your feedback about Neuron SDK! Let us know by taking the Neuron survey


It is recommended to deploy Neuron inside a preconfigured Deep Learning Container (DLC) from AWS. Running Neuron inside a container on Inf1 requires Docker version 18 (or newer) and a base AMI with aws-neuron-runtime-base and aws-neuron-dkms installed. It’s possible to also use a Neuron container on any instance type without the base and dkms package, but this is typically limited to compilation and development when running on instances without a Inf1 Device (inferentia). DLC images for Neuron can be obtained from here.

Documentation is organized based on the target deployment environment and use case. In most cases, it is recommended to use a preconfigured Deep Learning Container from AWS. Each DLC is pre-configured to have a recent version of Neuron components installed and is specific to the chosen ML Framework you want.

Release Notes

Resources for Neuron Runtime 1.x Users