← Back to community benchmarks

gemma-4-26b-a4b-it-mxfp8

M3 Ultra (60c) · 256 GB · 8bit · 2026-04-30

Performance

64k

tokens

1,210

PP tok/s

44.3

TG tok/s

54163

TTFT (ms)

29.3

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,237	79.0	25.1 GB	view
4k	1,593	78.5	25.7 GB	view
8k	1,653	72.1	25.9 GB	view
16k	1,618	68.4	26.3 GB	view
32k	1,475	58.7	27.3 GB	view
64k	1,210	44.3	29.3 GB	current