← Back to community benchmarks

Qwen3.5-4B

M4 (10c) · 16 GB · 4bit · 2026-04-29

Performance

4k

tokens

310.4

PP tok/s

34.9

TG tok/s

13197

TTFT (ms)

3.9

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 16 GB

GPU Cores 10

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	324.4	103.9	4.0 GB	view
4k	310.4	34.9	3.9 GB	current