← Back to community benchmarks

Qwen3.6-27B-MTPLX-Optimized

M5 Max (40c) · 128 GB · 4bit · 2026-05-09

Performance

16k

tokens

809.5

PP tok/s

25.3

TG tok/s

20240

TTFT (ms)

25.7

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9.dev1

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	698.8	26.0	21.8 GB	view
4k	851.1	26.7	23.2 GB	view
8k	817.1	26.4	24.2 GB	view
16k	809.5	25.3	25.7 GB	current
32k	717.5	24.1	28.7 GB	view
64k	595.1	21.5	34.8 GB	view
128k	427.4	17.4	47.1 GB	view