← Back to community benchmarks

Qwen3.6-27B-oQ6e

M5 Max (40c) · 128 GB · 6bit · 2026-05-05

Performance

8k

tokens

783.7

PP tok/s

23.0

TG tok/s

10453

TTFT (ms)

24.5

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	723.3	23.9	22.4 GB	view
4k	825.7	23.6	23.8 GB	view
8k	783.7	23.0	24.5 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	23.9	1.00×
2×	38.3	1.60×
4×	47.5	1.99×