← Back to community benchmarks

gemma-4-E4B-it-oQ4

M5 (10c) · 32 GB · 4bit · 2026-05-09

Performance

32k

tokens

1,313

PP tok/s

25.5

TG tok/s

24955

TTFT (ms)

5.8

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,275	36.9	5.0 GB	view
4k	1,589	36.7	5.0 GB	view
8k	1,645	34.9	5.2 GB	view
16k	1,500	30.4	5.3 GB	view
32k	1,313	25.5	5.8 GB	current
64k	1,096	18.3	6.6 GB	view
128k	615.0	7.8	8.4 GB	view