← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (20c) · 64 GB · 8bit · 2026-05-30

Performance

32k

tokens

1,699

PP tok/s

38.8

TG tok/s

19289

TTFT (ms)

27.5

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.12

macOS macOS 26.5

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
32k	1,699	38.8	27.5 GB	current
64k	1,343	30.4	30.0 GB	view
128k	936.1	21.2	34.2 GB	view
195k	691.9	16.2	35.9 GB	view