← Back to community benchmarks

Qwen3.5-4B

M3 Pro (14c) · 36 GB · 4bit · 2026-04-16

Performance

16k

tokens

451.2

PP tok/s

37.7

TG tok/s

36316

TTFT (ms)

5.2

Peak mem (GB)

Hardware

Chip M3 Pro (14c)

Memory 36 GB

GPU Cores 14

Software

oMLX v0.3.5

macOS macOS 26.3.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	447.1	48.8	3.4 GB	view
4k	478.6	46.5	4.1 GB	view
8k	474.9	43.4	4.5 GB	view
16k	451.2	37.7	5.2 GB	current
32k	402.9	30.3	6.7 GB	view
64k	332.3	17.9	9.7 GB	view
128k	236.2	9.6	16.1 GB	view