← Back to community benchmarks

gemma-4-12B-it

M1 Ultra (48c) · 64 GB · 8bit · 2026-06-05

Performance

128k

tokens

187.9

PP tok/s

21.1

TG tok/s

697435

TTFT (ms)

20.6

Peak mem (GB)

Hardware

Chip M1 Ultra (48c)

Memory 64 GB

GPU Cores 48

Software

oMLX v0.4.2.dev2

macOS macOS 26.5

Context 131,072

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
16k	278.6	29.9	13.9 GB	view
32k	257.6	27.9	15.1 GB	view
64k	233.0	25.3	17.0 GB	view
128k	187.9	21.1	20.6 GB	current