← Back to community benchmarks

Qwen3.6-27B-MTP-Optimized

M3 Max (40c) · 48 GB · 4bit · 2026-05-13

Performance

1k

tokens

210.2

PP tok/s

23.1

TG tok/s

4872

TTFT (ms)

15.8

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 48 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	210.2	23.1	15.8 GB	current
4k	216.3	22.2	17.3 GB	view
8k	198.3	21.9	18.3 GB	view
16k	201.9	20.3	19.8 GB	view
32k	170.2	15.1	22.8 GB	view