← Back to community benchmarks

Youssofal--Qwen3.6-27B-MTPLX-Optimized

M4 Max (40c) · 64 GB · 4bit · 2026-06-17

Performance

16k

tokens

239.4

PP tok/s

22.8

TG tok/s

68426

TTFT (ms)

23.6

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.4.4rc2

macOS macOS 26.5.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	238.5	24.5	19.6 GB	view
4k	247.8	24.2	21.1 GB	view
8k	246.0	23.7	22.1 GB	view
16k	239.4	22.8	23.6 GB	current
32k	225.8	21.4	26.6 GB	view
64k	199.9	18.9	32.6 GB	view