← Back to community benchmarks

Qwen3.6-27B

M4 Max (40c) · 48 GB · 4bit · 2026-04-24

Performance

32k

tokens

201.7

PP tok/s

21.1

TG tok/s

162452

TTFT (ms)

21.6

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	232.1	28.2	15.8 GB	view
4k	214.1	26.6	17.3 GB	view
8k	209.8	24.7	17.9 GB	view
16k	219.6	24.0	19.1 GB	view
32k	201.7	21.1	21.6 GB	current
64k	169.6	13.2	26.7 GB	view