← Back to community benchmarks

gemma-4-E2B-Heretic-Uncensored

M4 (10c) · 16 GB · 4bit · 2026-04-18

Performance

1k

tokens

1,386

PP tok/s

61.0

TG tok/s

739

TTFT (ms)

2.9

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.3.6

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,386	61.0	2.9 GB	current
4k	1,814	58.6	3.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	61.0	1.00×
2×	115.6	1.90×
4×	139.2	2.28×