← Back to community benchmarks

Qwen3.6-27B-3bit

M3 Max (30c) · 36 GB · 4bit · 2026-06-17
Performance
1k
tokens
157.8
PP tok/s
21.1
TG tok/s
6488
TTFT (ms)
13.2
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.4.4
macOS macOS 26.4
Context 1,024