← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M4 Max (40c) · 48 GB · 8bit · 2026-05-20

Performance

4k

tokens

204.3

PP tok/s

22.0

TG tok/s

20049

TTFT (ms)

30.3

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 48 GB

GPU Cores 40

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	237.2	26.9	28.8 GB	view
4k	204.3	22.0	30.3 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	26.9	1.00×
2×	28.5	1.06×
4×	36.5	1.36×