← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 24 GB · 8bit · 2026-05-19

Performance

1k

tokens

180.7

PP tok/s

11.3

TG tok/s

5668

TTFT (ms)

9.8

Peak mem (GB)

Hardware

Chip M4 (10c)

Memory 24 GB

GPU Cores 10

Software

oMLX v0.3.9.dev2

macOS macOS 26.4.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	180.7	11.3	9.8 GB	current
4k	178.8	11.2	10.4 GB	view
8k	174.1	10.7	11.0 GB	view
16k	141.2	10.0	11.8 GB	view