← Back to community benchmarks

Qwen3.5-9B-OptiQ

M2 Ultra (60c) · 128 GB · 4bit · 2026-03-10

Performance

4k

tokens

438.5

PP tok/s

58.8

TG tok/s

9341

TTFT (ms)

7.0

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 128 GB

GPU Cores 60

Software

oMLX v0.2.7

macOS macOS 26.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	399.4	59.9	6.8 GB	view
4k	438.5	58.8	7.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	59.9	1.00×
2×	82.2	1.37×
4×	96.7	1.61×