← Back to community benchmarks

gemma-4-E4B-it

M5 Pro (20c) · 24 GB · 8bit · 2026-05-13

Performance

128k

tokens

2,123

PP tok/s

24.5

TG tok/s

61736

TTFT (ms)

13.3

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 24 GB

GPU Cores 20

Software

oMLX v0.3.9.dev1

macOS macOS 26.5

Context 131,072

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	3,590	49.7	8.9 GB	view
4k	3,923	48.5	9.1 GB	view
8k	3,928	47.0	9.2 GB	view
16k	3,771	44.5	9.5 GB	view
32k	3,433	39.8	10.2 GB	view
64k	2,865	32.6	11.0 GB	view
128k	2,123	24.5	13.3 GB	current