Apple's M5 Ultra and M6 chips show significant GPU performance gains in early Geekbench 7 benchmarks, with the M5 Ultra achieving 59% higher Metal scores than the M3 Ultra and the M6 showing 34% improvement over the M5. New Mac Studio and Mac mini models featuring these chips launch Tuesday.
Apple's M6 Mac mini and competing mini PCs from Asus, Lenovo, and Acer are racing to become the primary devices for running AI models locally, with prices ranging from $900 to $3,700. These compact computers feature specialized AI accelerators and aim to challenge Apple's dominance, though clustering multiple machines may be necessary for meaningful on-device AI performance.
Apple's M6 chip benchmark results surfaced ahead of next week's Mac mini launch, showing approximately 12% single-core and 27% multi-core CPU performance gains over the M5 chip. The M6 achieves the fastest single-threaded CPU performance across all Macs and PCs in Geekbench 7, validating Apple's performance claims.
A Geekbench 6.3 benchmark of Apple's M6 Mac mini shows single-core and multi-core performance gains of roughly 10% and 20% respectively over the M5, aligning with Apple's claims. Surprisingly, the M6 trails Apple's A20 Pro smartphone chip in single-core performance and clock speed despite better thermal conditions, possibly due to different core design choices between the M-series and A-series.
A technical guide for running Gemma 4 26B on a Mac Mini M4 by raising the Metal GPU memory limit from the macOS default to enable full GPU utilization. The article explains why the model defaults to 66% CPU usage due to unified memory constraints and provides commands and configuration to achieve 100% GPU processing.