← Back to community benchmarks

Qwen3.6-27B-oQ4-fp16-mtp

M1 Max (32c) · 64 GB · 4bit · 2026-05-19

Performance

8k

tokens

149.1

PP tok/s

17.4

TG tok/s

54949

TTFT (ms)

18.5

Peak mem (GB)

Hardware

Chip M1 Max (32c)

Memory 64 GB

GPU Cores 32

Software

oMLX v0.3.8

macOS macOS 15.6.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	145.5	18.8	16.4 GB	view
4k	150.0	18.1	17.9 GB	view
8k	149.1	17.4	18.5 GB	current
16k	145.5	7.3	20.4 GB	view