← Back to community benchmarks

gemma-4-26b-a4b-it

M5 (10c) · 32 GB · 6bit · 2026-04-03

Performance

1k

tokens

794.6

PP tok/s

25.2

TG tok/s

1289

TTFT (ms)

20.9

Peak mem (GB)

Hardware

Chip M5 (10c)

Memory 32 GB

GPU Cores 10

Software

oMLX v0.3.1

macOS macOS 26.2

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	794.6	25.2	20.9 GB	current
4k	903.3	25.6	21.4 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	25.2	1.00×
2×	38.3	1.52×
4×	47.5	1.88×