← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 16 GB · 4bit · 2026-03-19

Performance

8k

tokens

210.3

PP tok/s

19.2

TG tok/s

38958

TTFT (ms)

7.7

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.2.19

macOS macOS 26.3.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	222.5	20.1	7.3 GB	view
8k	210.3	19.2	7.7 GB	current
16k	198.6	17.6	8.4 GB	view