← Back to community benchmarks

Qwen3.6-27B-oQ8-fp16-mtp

M2 Ultra (60c) · 192 GB · 8bit · 2026-05-15

Performance

16k

tokens

308.2

PP tok/s

25.5

TG tok/s

53155

TTFT (ms)

32.8

Peak mem (GB)

Hardware

Chip M2 Ultra (60c)

Memory 192 GB

GPU Cores 60

Software

oMLX v0.3.9.dev1

macOS macOS 26.3.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	291.2	27.6	28.8 GB	view
4k	313.7	27.5	30.3 GB	view
8k	315.1	26.9	31.3 GB	view
16k	308.2	25.5	32.8 GB	current
32k	291.3	23.9	35.8 GB	view