← Back to community benchmarks

gemma-4-e4b-it

M5 Pro (20c) · 48 GB · 8bit · 2026-04-14

Performance

1k

tokens

2,070

PP tok/s

49.4

TG tok/s

495

TTFT (ms)

8.0

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.5

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,070	49.4	8.0 GB	current
4k	3,287	47.3	8.1 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	49.4	1.00×
2×	92.1	1.86×
4×	139.6	2.83×