← Back to community benchmarks

Qwen3.6-27B-oQ6

M4 Max (40c) · 128 GB · 6bit · 2026-05-02
Performance
16k
tokens
805.9
PP tok/s
16.2
TG tok/s
20331
TTFT (ms)
27.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 223.9 101.9 23.9 GB view
4k 200.6 20.6 24.8 GB view
8k 224.6 19.4 25.8 GB view
16k 805.9 16.2 27.1 GB current
32k 793.3 16.2 28.4 GB view
64k 679.5 14.7 30.2 GB view