← Back to community benchmarks

Qwen3.5-9B

M2 Ultra (60c) · 64 GB · 4bit · 2026-03-12

Performance

4k

tokens

634.4

PP tok/s

88.3

TG tok/s

6456

TTFT (ms)

7.0

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 64 GB

GPU Cores 60

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	599.5	92.3	6.7 GB	view
4k	634.4	88.3	7.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	92.3	1.00×
2×	119.5	1.29×
4×	137.4	1.49×