← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 16 GB · 4bit · 2026-04-28

Performance

8k

tokens

220.3

PP tok/s

20.0

TG tok/s

37187

TTFT (ms)

7.4

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.3.6

macOS macOS 26.3

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	179.2	79.2	6.7 GB	view
4k	175.1	20.9	7.1 GB	view
8k	220.3	20.0	7.4 GB	current
16k	632.4	14.9	8.2 GB	view
32k	575.8	14.7	8.9 GB	view