← Back to community benchmarks

Qwen3.5-9B

M5 Max (32c) · 36 GB · 4bit · 2026-03-18

Performance

4k

tokens

2,301

PP tok/s

77.2

TG tok/s

1781

TTFT (ms)

7.4

Peak mem (GB)

Hardware

Chip M5 Max (32c)

Memory 36 GB

GPU Cores 32

Software

oMLX v0.2.19.dev1

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,510	80.3	6.7 GB	view
4k	2,301	77.2	7.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	80.3	1.00×
2×	146.9	1.83×
4×	202.7	2.52×