GEMMA 4

TurboFieldfare is a custom Swift and Metal inference engine that runs Google's 26B-parameter Gemma 4 MoE model in roughly 2 GB of RAM on any Apple Silicon Mac, including 8 GB base models.

Gemma 4 ships byte-for-byte open weights from Google DeepMind. How developers deploy it locally, fine-tune it, and ship agents on top of it.