Customers across diverse industries are defining entirely new categories of products and experiences by running intelligent applications that use ML at the core. These applications are becoming more expensive to run in production. AWS Inferentia is a custom-built machine learning inference chip designed to provide high throughput, low latency inference performance at an extremely low cost. Each chip provides hundreds of TOPS of inference throughput to allow complex models to make fast predictions. Join this session to see the latest developments using AWS Inferentia and how they can lower your inference costs in the future.
- AWS re:Invent 2019: [REPEAT 1] Deliver high performance ML inference with AWS Inferentia (CMP324-R1) ( Download)
- AWS re:Invent 2019: AWS infrastructure for large-scale training at Facebook AI (CMP304-R1) ( Download)
- AWS re:Invent 2020: Machine learning inference with Amazon EC2 Inf1 instances ( Download)
- AWS re:Invent 2019: Amazon.com automating machine learning deployments at scale (ARC340-R1) ( Download)
- EC2 15th Anniversary: Accelerate AI/ML adoption with AWS Inferentia | AWS Events ( Download)
- Deep Learning hardware acceleration with AWS Inferentia ( Download)
- AWS re:Invent 2020: How to choose the right instance type for ML inference ( Download)
- AWS re:Invent 2019: Deep dive on Arm-based EC2 instances powered by AWS Graviton (CMP322-R1) ( Download)
- AWS re:Invent 2019: Deep learning applications with TensorFlow, featuring Fannie Mae (AIM410-R1) ( Download)
- Amazon Alexa adopts Amazon EC2 Inf1 instances powered by AWS Inferentia ( Download)
- AWS Container Day - AWS Inferentia on Amazon EKS ( Download)
- Save up to 71% on Your Machine Learning Inference Costs Using Amazon EC2 Inf1 - AWS Online Tech Talk ( Download)
- AWS re:Invent 2021 - {New Launch} Introducing AWS Trainium-based Amazon EC2 Trn1 instances ( Download)
- AWS re:Invent 2019: Start using AI/ML/DL in your .NET applications today (WIN311) ( Download)
- AWS re:Invent 2021 - The journey of silicon innovation at AWS ( Download)