← Back to community benchmarks

Qwen36-27B-oQ4

M4 Max (40c) · 128 GB · 4bit · 2026-04-30
Performance
16k
tokens
240.4
PP tok/s
24.1
TG tok/s
68152
TTFT (ms)
19.6
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 15.7.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 227.4 84.1 17.6 GB view
4k 231.0 27.2 17.7 GB view
8k 247.0 26.2 18.4 GB view
16k 240.4 24.1 19.6 GB current
32k 226.9 21.9 22.1 GB view