The creators of the popular open-source framework PyTorch Lightning, Grid.ai has recently launched a new platform that enables you to train AI models on the cloud at scale. Named Grid, this new platform is extremely helpful for researchers, machine learning engineers and data scientists who want to do so without making any changes in the code.
What Do You Need to Know About Grid.ai?
Grid.ai helps companies to train their AI models on cloud GPUs and CPUs using their laptops. Their aim is to help simplify AI research. Found by the creator of PyTorch Lightning, William Falcon, the company has raised $18.6 million as venture capital from Index Ventures, Bain Capital Ventures and Firstminute.
What Is Grid And How Does It Benefit You?
To be specific, Grid is a web application that provides an interface to train AI models on GPUs and processors. It computes in real-time and works at the level of production which helps in quantifying the R&D efforts. By gaining access to Jupyter notebooks, you can additionally bundle your answers with the python code with which you produced it.
In order to use the app, you’ll just have to choose a plan – individual, team or enterprise. The costing varies as it depends on the cost of cloud processing. Individual plans cost you 20% in addition to the cloud cost while team plans cost $1,000 per month and 15% more in addition to the cloud cost.
Once you’ve subscribed, you can make use of the flexible command line interface to test your AI models. Since you don’t spend your time and money on developing the infrastructure setup, your research and production speed accelerates dramatically.