← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M4 Max (40c) · 128 GB · 4bit · 2026-05-22

Performance

16k

tokens

238.9

PP tok/s

35.6

TG tok/s

68572

TTFT (ms)

20.0

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9

macOS macOS 26.5

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	239.1	38.7	16.7 GB	view
4k	247.9	39.3	18.2 GB	view
8k	245.8	37.5	18.8 GB	view
16k	238.9	35.6	20.0 GB	current
32k	222.9	30.9	22.5 GB	view
64k	195.7	26.1	27.6 GB	view
128k	155.3	15.8	37.9 GB	view