AI with Apple MLX

Build AI apps with the power of open source Apple MLX framework.

hot stack

High-performance AI

Built for Apple silicon chips

MLX is already two times faster than PyTorch, fully optimized for low memory conditions, unified memory and large models. Get up to 100 tokens per second on M2 Ultra configurations.

Fully open source

Built by Apple research team with an MIT license

Learn more...

Works with C, Python and Swift

MLX is already optmized for C, Python and Swift languages. Easy to use: "Brew install mlx"

Learn more...

Ready for inference

MLX server is an open source project to take advantage of the framework for easy inference

Learn more...


Options   Documentation

MacWeb Computing is an independent company not affiliated with Apple


Get a Mac in the cloud

We can help

Request a quote   Request assistance