Modular AI Engine
The world’s fastest unified inference engine, supercharging any model from TensorFlow or PyTorch on a wide range of hardware.
We’re excited to share an early look at the Modular compute and AI infrastructure stack with the following documentation. Our mission to revolutionize the AI developer experience still has a ways to go, but we’re making rapid progress.
The AI Engine isn’t generally available yet, but you can sign up here for early access.
And although the Mojo language is still young, it’s now available with the Mojo SDK!