← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M5 Pro (20c) · 48 GB · 4bit · 2026-05-14

Performance

16k

tokens

450.4

PP tok/s

24.2

TG tok/s

36379

TTFT (ms)

20.7

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.9.dev2

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
4k	469.3	26.5	18.2 GB	view
8k	468.7	25.3	19.2 GB	view
16k	450.4	24.2	20.7 GB	current
32k	411.5	22.1	23.7 GB	view
64k	342.2	17.9	29.7 GB	view