← Back to community benchmarks

Qwen3.6-27B-oQ4

M4 Max (40c) · 64 GB · 4bit · 2026-04-23
Performance
32k
tokens
120.6
PP tok/s
13.9
TG tok/s
271768
TTFT (ms)
22.2
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 238.4 27.5 16.4 GB view
4k 200.6 24.4 17.9 GB view
8k 188.2 22.9 18.5 GB view
16k 160.0 11.3 19.7 GB view
32k 120.6 13.9 22.2 GB current
Batching Results
Batch Size TG tok/s Speedup
27.5 1.00×
37.2 1.35×
78.8 2.87×