← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M4 Max (40c) · 128 GB · 8bit · 2026-05-16

Performance

8k

tokens

156.8

PP tok/s

15.5

TG tok/s

52256

TTFT (ms)

31.3

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9.dev2

macOS macOS 15.7

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	142.4	16.0	28.8 GB	view
4k	158.4	16.6	30.3 GB	view
8k	156.8	15.5	31.3 GB	current