← Back to community benchmarks

Gemma-4-26B-A4B

M3 Pro (14c) · 18 GB · 3bit · 2026-04-17

Performance

1k

tokens

463.6

PP tok/s

45.1

TG tok/s

2209

TTFT (ms)

11.4

Peak mem (GB)

Hardware

Chip M3 Pro (14c)

Memory 18 GB

GPU Cores 14

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 1,024

Batching Results

Batch Size	TG tok/s	Speedup
1×	45.1	1.00×
2×	62.8	1.39×