← Back to community benchmarks

Qwen3.6-27B

M5 Max (40c) · 128 GB · 8bit · 2026-05-18

Performance

8k

tokens

768.1

PP tok/s

15.2

TG tok/s

10666

TTFT (ms)

35.6

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.8

macOS macOS 26.5

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	606.9	15.6	33.0 GB	view
4k	727.3	15.4	34.6 GB	view
8k	768.1	15.2	35.6 GB	current
16k	735.9	14.9	37.1 GB	view
32k	641.2	14.3	40.1 GB	view
64k	488.8	12.5	46.2 GB	view