# Using AWS Deep Learning Containers on Amazon ECS
<a name="deep-learning-containers"></a>

AWS Deep Learning Containers provide a set of Docker images for training and serving models in TensorFlow and Apache MXNet (Incubating) on Amazon ECS. Deep Learning Containers enable optimized environments with TensorFlow, NVIDIA CUDA (for GPU instances), and Intel MKL (for CPU instances) libraries. Container images for Deep Learning Containers are available in Amazon ECR to reference in Amazon ECS task definitions. You can use Deep Learning Containers along with Amazon Elastic Inference to lower your inference costs.

To get started using Deep Learning Containers without Elastic Inference on Amazon ECS, see [ Amazon ECS setup](https://docs.aws.amazon.com/deep-learning-containers/latest/devguide/deep-learning-containers-ecs-setup.html) in the *AWS Deep Learning AMIs Developer Guide*.