← Back to community benchmarks

Qwen3.5-4B-OptiQ

M3 Pro (18c) · 18 GB · 4bit · 2026-03-09

Performance

4k

tokens

510.4

PP tok/s

27.5

TG tok/s

8025

TTFT (ms)

5.4

Peak mem (GB)

Hardware

Chip M3 Pro (18c)

Memory 18 GB

GPU Cores 18

Software

oMLX v0.2.6

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	458.6	28.4	5.1 GB	view
4k	510.4	27.5	5.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	28.4	1.00×
2×	54.3	1.91×
4×	94.3	3.32×