← Back to community benchmarks

Qwen3.6-27B-5bit

M4 Max (40c) · 48 GB · 5bit · 2026-07-12
Performance
32k
tokens
221.0
PP tok/s
21.4
TG tok/s
148241
TTFT (ms)
24.8
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 48 GB
GPU Cores 40
Software
oMLX v0.4.4
macOS macOS 26.5.2
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
32k 221.0 21.4 24.8 GB current
64k 190.9 18.6 25.4 GB view