← Back to community benchmarks

gemma-4-e4b-it-mxfp8

M4 Pro (20c) · 48 GB · 8bit · 2026-04-17

Performance

4k

tokens

1,101

PP tok/s

35.9

TG tok/s

3720

TTFT (ms)

7.9

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.6

macOS macOS 26.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	873.4	38.8	7.8 GB	view
4k	1,101	35.9	7.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	38.8	1.00×
2×	65.1	1.68×
4×	71.3	1.84×