Ollama adds MLX support on Apple silicon Macs

TL;DR:

• Ollama adopted MLX on March 31, 2026 to improve local AI model performance on Apple silicon Macs.
• The change applies to running AI models locally on Macs with Apple silicon.
• 9to5Mac reported the update as a performance improvement for Mac users running Ollama.

Ollama adopted MLX on March 31, 2026 to improve local AI model performance on Apple silicon Macs. 9to5Mac reported that the update makes Ollama faster for users who run AI models locally on Mac hardware.

The report identifies Ollama as a tool for running AI models locally on a Mac and says it now uses MLX on Apple silicon Macs. The source excerpt does not provide benchmark numbers, supported model names, macOS version requirements, or a list of compatible Mac devices beyond Apple silicon.

This update fits a broader shift toward on-device AI on Apple hardware, where developers use Apple silicon optimizations to run models locally instead of in the cloud. The source does not mention eSIM, operators, devices with eUICC, or GSMA standards such as SGP.32, the GSMA remote SIM provisioning specification for IoT.

Related Questions

Does Ollama use MLX on Apple silicon Macs?

Yes. 9to5Mac reported on March 31, 2026 that Ollama adopted MLX for faster AI performance on Apple silicon Macs.

What changed for Ollama on Mac?

Ollama added MLX support for Apple silicon Macs. The reported effect is faster local AI model performance on supported Mac hardware.