The Deep Learning AMI is a base Windows image provided by Amazon Web Services for use on Amazon Elastic Compute Cloud (Amazon EC2). It is configured with NVidia CUDA 8 and 9, SciPy, Conda and NVidia Driver 385.54. The AMI also has MXNet, Caffe and TensorFlow. The AMI is specially designed to provide high performance execution environment for deep learning on EC2 Accelerated Computing instances. Last but not the least, it also includes Anaconda Data Science Platform for Python2 and Python3. The Deep Learning AMI is provided at no additional charge to Amazon EC2 users.

Release tags/Branches used:
MXNet 0.12.0 Release Candidate
Caffe /windows branch commit #5854
TensorFlow 1.4

Check the AMI release notes for more details: http://docs.aws.amazon.com/dlami/latest/devguide/appendix-ami-release-notes.html.

Please note: Microsoft Windows Server 2012 and 2012 R2 are reaching end of support on October 10th 2023. These images will be removed from AWS Marketplace by October 10th, and you will no longer be able to access license included AMIs provided by Amazon after this date. Custom AMIs you own and active EC2 instances will not be impacted, but we recommend upgrading your current workloads. Please review the end of support FAQ here: https://aws.amazon.com/windows/faq/#eos-microsoft-products. Amazon EC2 running Microsoft Windows Server is a fast and dependable environment for deploying applications using the Microsoft Web Platform. Amazon EC2 enables you to run any compatible Windows-based solution on AWS' high-performance, reliable, cost-effective, cloud computing platform. Common Windows use cases include Enterprise Windows-based application hosting, website and web-service hosting, data processing, media transcoding, distributed testing, ASP.NET application hosting, and any other application requiring Windows software.