← Back to community benchmarks

gemma-4-26b-a4b-it-oQ8

M3 Ultra (60c) · 96 GB · 8bit · 2026-04-30

Performance

1k

tokens

1,313

PP tok/s

81.4

TG tok/s

780

TTFT (ms)

25.8

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,313	81.4	25.8 GB	current
4k	1,658	77.6	26.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	81.4	1.00×
2×	125.9	1.55×
4×	169.1	2.08×