← Back to community benchmarks

gemma-4-26B-A4B-it

M1 Ultra (64c) · 128 GB · 8bit · 2026-04-17

Performance

1k

tokens

686.2

PP tok/s

60.2

TG tok/s

1492

TTFT (ms)

25.7

Peak mem (GB)

Hardware

Chip M1 Ultra (64c)

Memory 128 GB

GPU Cores 64

Software

oMLX v0.3.5

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	686.2	60.2	25.7 GB	current
4k	813.7	56.4	26.2 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	60.2	1.00×
2×	88.0	1.46×
4×	123.3	2.05×