The AWS Deep Learning AMIs are a quick and easy way to get started with machine learning. The AMIs with Source Code now come with TensorFlow 1.5.0-rc1 which supports NVidia CUDA 9 and cuDNN 7 drivers that take advantage of the V100 Volta GPUs powering the EC2 P3 instances. The Conda-based Deep Learning AMIs now come with the latest framework versions of Caffe, Keras 2.1.3, Microsoft Cognitive Toolkit 2.3.1, and Theano 1.0. The Conda-based AMIs also include TensorBoard for monitoring and debugging a TensorFlow model, and TensorFlow Serving to export TensorFlow models and create a server running a gRPC service for serving the model. Apache MXNet users can now benefit from the MXNet Model Server to quickly deploy an HTTP-based inference API for their models.

Source: Amazon Web Services News AWS Deep Learning AMIs now come with TensorFlow 1.5.0 and new Model Serving Capabilities