← Back to community benchmarks

gemma-4-26b-a4b-it

M3 Pro (18c) · 36 GB · 4bit · 2026-04-17

Performance

32k

tokens

453.3

PP tok/s

10.3

TG tok/s

72283

TTFT (ms)

16.0

Peak mem (GB)

Hardware

Chip M3 Pro (18c)

Memory 36 GB

GPU Cores 18

Software

oMLX v0.3.5

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	510.0	36.2	14.2 GB	view
4k	540.2	31.8	14.7 GB	view
8k	547.4	33.6	14.8 GB	view
16k	502.5	26.5	15.2 GB	view
32k	453.3	10.3	16.0 GB	current
64k	360.7	6.4	17.7 GB	view
128k	223.5	4.2	21.2 GB	view
195k	183.4	2.6	25.0 GB	view