← Back to community benchmarks

Qwen3.6-27B-oQ6-fp16-mtp

M5 Max (40c) · 128 GB · 6bit · 2026-06-04

Performance

8k

tokens

723.7

PP tok/s

27.0

TG tok/s

11320

TTFT (ms)

25.0

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.4.1

macOS macOS 26.5

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	723.7	27.0	25.0 GB	current
16k	594.1	25.5	26.9 GB	view
32k	500.5	17.3	29.9 GB	view
64k	228.9	10.0	36.0 GB	view