Dan Woods demonstrates running a 397B parameter AI model locally on a MacBook Pro, using Apple’s flash-based method to reduce memory use and enable large-model inference.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results