OctoML and AWS Better Together

OctoML is a SaaS solution that automates and optimizes trained ML models for seamless deployment on a myriad of AWS hardware targets including NVIDIA GPUs, Intel and AMD CPUs and AWS Graviton. OctoML’s model-as-function packaging enables ML models to be integrated into the same AWS workflows as the rest of a customer's application stack.

AWS Partner Network logo

Benefits of OctoML and AWS

Accelerate Deployment
Accelerate Deployment
Accelerate Deployment

Machine Learning (ML) for ML that chooses the best production packaging from among thousands of combinations–shrinking model deployment times to hours.

Increase Success
Increase Success
Increase Success

Automating away error-prone human tasks leads to predictable and stable production outcomes for deployment on AWS EKS, ECS, or SageMaker.

Make Accessible
Make Accessible
Make Accessible

Creating models-as-functions empowers your developer and Ops teams to leverage DevOps best practices rather than learn a new discipline.

OctoML SaaS Platform

OctoML Platform overview diagram

OctoML and AWS Reference Architecture

OctoML in the AWS ML lifecycle diagram

Accelerate Your AI Innovation