← Back to community benchmarks

gemma-4-E4B-it-oQ8e

M5 (10c) · 32 GB · 8bit · 2026-05-21

Performance

32k

tokens

1,585

PP tok/s

34.3

TG tok/s

20673

TTFT (ms)

9.4

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.3.9

macOS macOS 26.5

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,882	58.0	8.4 GB	view
4k	1,973	54.6	8.4 GB	view
8k	1,992	50.9	8.6 GB	view
16k	1,849	44.1	8.9 GB	view
32k	1,585	34.3	9.4 GB	current
64k	1,325	23.8	10.6 GB	view
128k	940.3	13.5	13.0 GB	view