← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M4 Max (40c) · 128 GB · 8bit · 2026-05-24

Performance

1k

tokens

235.2

PP tok/s

26.7

TG tok/s

4353

TTFT (ms)

28.8

Peak mem (GB)

Hardware

Chip M4 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.9

macOS macOS 15.7.4

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	235.2	26.7	28.8 GB	current
4k	238.7	26.2	30.3 GB	view
64k	188.2	18.5	39.7 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	26.7	1.00×
2×	29.9	1.12×
4×	41.3	1.55×