← Back to community benchmarks

nightmedia:Qwen3.6-27B-Holodeck-Qwopus-GLM-qx64-hi

M5 Max (40c) · 128 GB · 6bit · 2026-06-14

Performance

4k

tokens

905.5

PP tok/s

28.0

TG tok/s

4523

TTFT (ms)

20.5

Peak mem (GB)

Hardware

Chip M5 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.4.4rc1

macOS macOS 26.5.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	757.9	29.1	19.1 GB	view
4k	905.5	28.0	20.5 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	29.1	1.00×
2×	49.3	1.69×
4×	109.2	3.75×