← Back to community benchmarks

Qwen3.6-27B-uncensored-heretic-v2-Native-MTP-Preserved-oQ4-mtp

M1 Max (32c) · 64 GB · 4bit · 2026-05-13

Performance

64k

tokens

56.5

PP tok/s

7.8

TG tok/s

1159139

TTFT (ms)

29.7

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.3.9.dev2

macOS macOS 26.3

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
8k	66.0	10.8	18.8 GB	view
32k	62.5	10.2	23.2 GB	view
64k	56.5	7.8	29.7 GB	current