← Back to community benchmarks

Qwen3.5-9B

M3 Pro (18c) · 18 GB · 4bit · 2026-03-10

Performance

1k

tokens

345.4

PP tok/s

27.7

TG tok/s

2965

TTFT (ms)

6.7

Peak mem (GB)

Hardware

Chip M3 Pro (18c)

Memory 18 GB

GPU Cores 18

Software

oMLX v0.2.6

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	345.4	27.7	6.7 GB	current
4k	347.9	26.4	7.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	27.7	1.00×
2×	53.4	1.93×
4×	80.7	2.91×