← Back to community benchmarks

Qwen3.5-4B

M4 (10c) · 16 GB · 4bit · 2026-04-18

Performance

4k

tokens

309.1

PP tok/s

35.1

TG tok/s

13250

TTFT (ms)

3.9

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.3.6

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	291.7	105.8	4.3 GB	view
4k	309.1	35.1	3.9 GB	current
8k	337.4	32.6	4.3 GB	view