← Back to community benchmarks

Qwen3.6-27B-UD

M5 Max (40c) · 128 GB · 4bit · 2026-04-26
Performance
16k
tokens
757.2
PP tok/s
18.9
TG tok/s
21637
TTFT (ms)
28.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 771.8 21.2 25.2 GB view
4k 911.1 20.7 26.6 GB view
8k 860.4 20.1 27.3 GB view
16k 757.2 18.9 28.5 GB current
32k 669.3 17.1 31.0 GB view
64k 560.2 14.4 36.1 GB view