← Back to community benchmarks

Qwen3.5-9B

M3 Ultra (60c) · 256 GB · 4bit · 2026-04-15

Performance

4k

tokens

755.3

PP tok/s

105.8

TG tok/s

5423

TTFT (ms)

6.5

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.3.5

macOS macOS 26.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	847.0	269.7	8.2 GB	view
4k	755.3	105.8	6.5 GB	current
8k	1,109	100.3	6.9 GB	view
16k	1,086	90.5	7.6 GB	view
32k	1,021	74.0	9.1 GB	view
64k	898.0	54.5	12.2 GB	view