← Back to community benchmarks

Qwen3.6-27B

M5 Max (40c) · 128 GB · 4bit · 2026-04-25

Performance

16k

tokens

712.0

PP tok/s

29.8

TG tok/s

23011

TTFT (ms)

19.1

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	818.3	41.3	19.5 GB	view
4k	680.9	32.1	17.3 GB	view
8k	831.0	31.6	17.9 GB	view
16k	712.0	29.8	19.1 GB	current
32k	570.1	8.3	21.6 GB	view