← Back to community benchmarks

Qwen3.6-27B-oQ6-mtp

M1 Ultra (64c) · 128 GB · 6bit · 2026-05-19

Performance

8k

tokens

164.7

PP tok/s

20.8

TG tok/s

49744

TTFT (ms)

25.1

Peak mem (GB)

Hardware

Chip M1 Ultra (64c)

Memory 128 GB

GPU Cores 64

Software

oMLX v0.3.9rc1

macOS macOS 26.5

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	154.7	22.2	22.6 GB	view
4k	164.9	21.4	24.0 GB	view
8k	164.7	20.8	25.1 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	22.2	1.00×
2×	28.3	1.27×
4×	32.2	1.45×