← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M4 Max (40c) · 128 GB · 8bit · 2026-05-19

Performance

195k

tokens

114.4

PP tok/s

12.5

TG tok/s

1748973

TTFT (ms)

68.9

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9.dev2

macOS macOS 15.7.7

Context 200,000

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
64k	177.8	18.1	39.7 GB	view
128k	136.6	14.8	54.2 GB	view
195k	114.4	12.5	68.9 GB	current