← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · bf16 · 2026-05-25

Performance

8k

tokens

1,770

PP tok/s

38.7

TG tok/s

4629

TTFT (ms)

48.6

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.10

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,104	42.7	47.8 GB	view
4k	1,511	43.5	48.5 GB	view
8k	1,770	38.7	48.6 GB	current
16k	1,743	34.9	49.1 GB	view
32k	1,484	29.4	50.0 GB	view
64k	1,047	20.9	52.0 GB	view