← Back to community benchmarks

Qwen3.6-27B-oQ6-mtp

M1 Max (32c) · 64 GB · 6bit · 2026-05-23

Performance

32k

tokens

70.4

PP tok/s

10.5

TG tok/s

465332

TTFT (ms)

25.8

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.3.9

macOS macOS 15.7.4

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	63.8	10.3	22.5 GB	view
16k	71.2	10.7	23.9 GB	view
32k	70.4	10.5	25.8 GB	current
64k	66.8	8.4	28.9 GB	view