← Back to community benchmarks

Qwen3.6-27B-MTPLX-Optimized

M4 Max (40c) · 128 GB · 4bit · 2026-05-06

Performance

8k

tokens

209.7

PP tok/s

23.1

TG tok/s

39057

TTFT (ms)

22.8

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	230.7	52.5	23.3 GB	view
4k	207.3	23.8	21.8 GB	view
8k	209.7	23.1	22.8 GB	current