← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 16 GB · 4bit · 2026-03-12

Performance

4k

tokens

223.9

PP tok/s

19.8

TG tok/s

18290

TTFT (ms)

6.8

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.2.6

macOS macOS 15.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	223.9	19.8	6.8 GB	current
8k	214.6	17.3	7.0 GB	view