← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (16c) · 48 GB · 4bit · 2026-05-06

Performance

4k

tokens

1,630

PP tok/s

69.2

TG tok/s

2513

TTFT (ms)

14.9

Peak mem (GB)

Hardware

Chip M5 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,374	73.7	14.3 GB	view
4k	1,630	69.2	14.9 GB	current
64k	1,277	38.6	17.7 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	73.7	1.00×
2×	101.7	1.38×
4×	121.5	1.65×