← Back to community benchmarks

Qwopus3.6-27B-v2-oQ4-mtp

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

Performance

16k

tokens

236.5

PP tok/s

35.3

TG tok/s

69271

TTFT (ms)

20.7

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	238.0	38.9	16.7 GB	view
4k	246.1	39.9	18.2 GB	view
8k	244.0	37.0	19.2 GB	view
16k	236.5	35.3	20.7 GB	current
32k	220.0	31.3	23.7 GB	view
64k	196.1	25.3	29.7 GB	view