← Back to community benchmarks

Qwen3.6-27B

M2 Ultra (60c) · 64 GB · 4bit · 2026-04-26

Performance

16k

tokens

850.5

PP tok/s

31.5

TG tok/s

19264

TTFT (ms)

19.8

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 64 GB

GPU Cores 60

Software

oMLX v0.3.6

macOS macOS 26.3.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	171.2	54.7	18.7 GB	view
4k	160.3	29.4	17.5 GB	view
8k	188.8	27.6	18.1 GB	view
16k	850.5	31.5	19.8 GB	current
32k	838.5	30.4	21.0 GB	view