← Back to community benchmarks

gemma-4-26b-a4b-it

M1 Ultra (48c) · 128 GB · 8bit · 2026-04-05

Performance

8k

tokens

665.2

PP tok/s

32.1

TG tok/s

12315

TTFT (ms)

27.4

Peak mem (GB)

Hardware

Chip M1 Ultra (48c)

Memory 128 GB

GPU Cores 48

Software

oMLX v0.3.2

macOS macOS 15.7.2

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	697.8	34.1	26.8 GB	view
4k	691.4	35.0	27.3 GB	view
8k	665.2	32.1	27.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	34.1	1.00×
2×	55.1	1.62×
4×	82.5	2.42×