Hear from CIOs, CTOs, and other C-level and senior execs on data and AI strategies at the Future of Work Summit this January 12, 2022. Learn more
Today at TVMcon 2021, OctoML announced the release of its machine learning (ML) deployment platform, a solution designed to enable organizations to automate and scale deploying ML models to hardware and cloud infrastructure.
The platform is designed to deploy models to a range of cloud services and hardware infrastructure, including AWS, Microsoft Azure, Google Cloud Platform, NVIDIA GPUs, Intel CPUs, AMD CPUS, and edge platforms like NVIDIA Jetson and Arm Cortex-A.
OctoML’s machine learning platform is an edge AI optimization solution, a category of solutions designed to optimize the performance of AI solutions at the network’s edge with automated deployment so engineers don’t have to spend hours on administration and manual optimization.
Automated model deployment and optimal performance also mean that decision-makers can generate insights faster, which is traditionally very difficult to do manually due to the complexity of the models being processed.
Scaling machine learning models
One of the biggest challenges facing organizations using AI to derive insights, is the complexity of deploying and processing machine learning models.
“Enterprises today face significant challenges with scaling the deployment of their trained models. In fact, research shows that nearly two-thirds of models take over a month to deploy into production,” said Luis Ceze, CEO, OctoML.
“This is because model performance tuning and optimization is largely done manually. Also, models, software platforms, and inference targets are rapidly evolving, requiring highly skilled resources on an ongoing basis,” he said.
It’s a challenge that the organization aims to confront with the new Machine Learning deployment platform. This latest iteration breaks these bottlenecks, making machine learning economically viable and enabling faster innovation,” he said.
The solution uses TVM performance enhancements to process popular machine learning models up to twice as fast, so that organizations can generate insights faster.
The growth of edge AI
The announcement comes after OctoML raised $85 million in a Series C funding round led by Tiger Global Management, reaching total funding of $132 million.
The company’s growth is directly linked to the growth of the edge AI software market, which research estimates will reach a value of $2271.73 million by 2027, as more organizations move to the cloud and look for solutions that enable them to process data created at the network’s edge.
OctoML is one of a growing number of vendors providing organizations with solutions to help automate and optimize the deployment of machine learning models to generate insights at the network’s edge, with competitors ranging from Neural Magic, to CoCoPie, NeuReality, and DeepCube.
Neural Magic, one of OctoML’s main competitors, a vendor that offers open source modeling capabilities and software designed to deploy deep learning models in edge environments, recently announced closing a $30 million Series A funding round.
Another, known as CoCoPie, offers organizations a solution for optimizing AI models for edge devices, earlier this year raising $6 million in Series A funding, and achieving a valuation of $50 million.
OctoML is trying to differentiate itself from competitors by building a cohesive ecosystem of hardware and cloud partners so that organizations can optimize model performance in whatever cloud or hybrid cloud environment they’re working within. The end goal is to generate a solution that enables organizations to generate insights and eliminates some intensive manual labor of deploying them at the network’s edge.