← Back to community benchmarks

Qwen3.5-9B-OptiQ

M1 Pro (14c) · 16 GB · 4bit · 2026-05-15

Performance

1k

tokens

136.4

PP tok/s

25.2

TG tok/s

7508

TTFT (ms)

7.7

Peak mem (GB)

Hardware

Chip M1 Pro (14c)

Memory 16 GB

GPU Cores 14

Software

oMLX v0.3.9.dev2

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	136.4	25.2	7.7 GB	current
4k	140.2	24.5	8.4 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	25.2	1.00×
2×	32.8	1.30×
4×	34.9	1.38×