← Back to community benchmarks

Qwen3.6-27B-MXFP4-MTP

M3 Max (30c) · 36 GB · 4bit · 2026-05-21

Performance

4k

tokens

161.6

PP tok/s

14.1

TG tok/s

25344

TTFT (ms)

17.3

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 36 GB

GPU Cores 30

Software

oMLX v0.3.9rc1

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	151.5	14.4	15.9 GB	view
4k	161.6	14.1	17.3 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	14.4	1.00×
2×	34.6	2.40×
4×	42.6	2.96×