← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (20c) · 64 GB · 4bit · 2026-05-12

Performance

4k

tokens

1,791

PP tok/s

70.0

TG tok/s

2287

TTFT (ms)

14.9

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.9.dev1

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,488	73.0	14.3 GB	view
4k	1,791	70.0	14.9 GB	current
8k	2,073	66.2	15.1 GB	view
16k	1,803	58.6	15.6 GB	view
32k	1,537	49.0	16.5 GB	view
64k	1,224	36.0	18.5 GB	view