← Back to community benchmarks

Qwen3.5-9B

M4 Pro (20c) · 64 GB · 4bit · 2026-04-17

Performance

16k

tokens

395.5

PP tok/s

40.3

TG tok/s

41425

TTFT (ms)

7.6

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	418.7	49.3	5.8 GB	view
4k	442.7	48.0	6.5 GB	view
8k	439.2	45.1	6.9 GB	view
16k	395.5	40.3	7.6 GB	current
32k	365.2	34.8	9.1 GB	view
64k	340.3	26.3	12.2 GB	view
128k	279.9	17.8	18.5 GB	view
195k	228.8	12.8	25.1 GB	view