← Back to community benchmarks

Qwen3.6-27B-oQ4-fp16-mtp

M3 Max (40c) · 64 GB · 4bit · 2026-06-04

Performance

16k

tokens

198.4

PP tok/s

25.2

TG tok/s

82595

TTFT (ms)

20.7

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.10

macOS macOS 15.7.7

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	221.3	32.2	16.7 GB	view
4k	226.0	29.7	18.2 GB	view
8k	205.4	31.2	19.2 GB	view
16k	198.4	25.2	20.7 GB	current
32k	143.4	14.2	23.7 GB	view