Fact · Speech recognition
MLX, Apple's array framework, keeps its arrays in shared memory, so an operation can run on the CPU or the GPU without first copying the data between them.
Checked against its source on October 17, 2026
In the source's own words
“Arrays in MLX live in shared memory. Operations on MLX arrays can be performed on any of the supported device types without transferring data.”
MLX: README (ml-explore/mlx, GitHub), Apple machine learning research, GitHub (ml-explore/mlx), 2026, Key features, Unified memory bullet.
Open the original- What it measures
- The MLX README calls this its unified memory model. MLX Whisper is built on MLX, so Whisper models run through it use this memory model.
- What it does not show
- The README describes MLX in general. It does not give memory figures for Whisper specifically, so measure the memory your own model uses.