← Back to community benchmarks

Qwen3.5-9B

M4 Pro (16c) · 24 GB · 4bit · 2026-03-12

Performance

1k

tokens

354.6

PP tok/s

49.4

TG tok/s

2888

TTFT (ms)

6.7

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 24 GB

GPU Cores 16

Software

oMLX v0.2.8

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	354.6	49.4	6.7 GB	current
4k	361.8	47.5	7.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	49.4	1.00×
2×	78.6	1.59×
4×	75.5	1.53×