Built different
Other AI infrastructure is assembled from parts that weren’t designed to work together: one tool for serving, one for model optimization, and one (or more) for GPU kernels. Modular is built different. Modular is a unified stack, from the kernels to the cloud. This gives us more flexibility to deploy across NVIDIA, AMD, and other hardware, which gives you more options to get the performance you need at the right cost. It’s all built upon two key open source components: MAX and Mojo.MAX 🧑🚀
An open and extensible framework for building and serving AI.It’s the serving and modeling framework that makes our models performant
and portable. You can use it yourself to build and deploy your own models.
Mojo 🔥
An open source programming language built for the AI era.It’s a Pythonic and safe systems language that can write blazing fast code
for CPUs, GPUs, and ASICs. It powers all of our kernels, and you can use it
to accelerate any project.
