← Back to community benchmarks

Qwen3.6-27B-MXFP8-MTP

M4 Max (40c) · 64 GB · unknown · 2026-06-15

Performance

1k

tokens

209.5

PP tok/s

16.7

TG tok/s

4887

TTFT (ms)

28.5

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.4.4rc1

macOS macOS 26.5.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	209.5	16.7	28.5 GB	current
4k	223.5	16.4	30.6 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	16.7	1.00×
2×	25.9	1.55×
4×	27.7	1.66×