← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M3 Max (40c) · 64 GB · 8bit · 2026-05-20

Performance

4k

tokens

201.5

PP tok/s

12.7

TG tok/s

20331

TTFT (ms)

29.8

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 64 GB

GPU Cores 40

Software

oMLX v0.3.9rc1

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	205.2	12.7	28.3 GB	view
4k	201.5	12.7	29.8 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	12.7	1.00×
2×	25.2	1.98×
4×	50.9	4.01×