← Back to community benchmarks

Qwen3.6-27B

M5 Max (40c) · 128 GB · 8bit · 2026-06-15

Performance

1k

tokens

690.8

PP tok/s

18.6

TG tok/s

1482

TTFT (ms)

28.3

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.4.4rc1

macOS macOS 26.5

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	690.8	18.6	28.3 GB	current
4k	714.9	18.5	29.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	18.6	1.00×
2×	30.2	1.62×
4×	40.9	2.20×