← Back to community benchmarks

Qwen3.6-27B

M4 Pro (20c) · 64 GB · 4bit · 2026-05-04
Performance
8k
tokens
110.9
PP tok/s
14.3
TG tok/s
73880
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.3
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 116.6 26.4 17.2 GB view
4k 110.6 14.4 17.3 GB view
8k 110.9 14.3 18.3 GB current
16k 110.7 13.5 19.8 GB view