← Back to community benchmarks

Qwen3.5-9B

M5 Max (40c) · 128 GB · 4bit · 2026-05-15

Performance

8k

tokens

2,240

PP tok/s

90.1

TG tok/s

3657

TTFT (ms)

7.0

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9.dev2

macOS macOS 26.5

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,809	93.8	5.8 GB	view
4k	2,043	94.6	6.5 GB	view
8k	2,240	90.1	7.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	93.8	1.00×
2×	163.1	1.74×
4×	206.9	2.21×