← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M3 Max (40c) · 128 GB · 4bit · 2026-05-10
User commentNATIVE MTP ON
Performance
8k
tokens
194.2
PP tok/s
23.2
TG tok/s
42188
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 15.7.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 189.2 21.2 16.8 GB view
4k 197.8 26.9 18.2 GB view
8k 194.2 23.2 19.2 GB current
16k 133.8 8.3 20.7 GB view
Batching Results
Batch Size TG tok/s Speedup
21.2 1.00×
12.2 0.58×
54.3 2.56×
70.5 3.33×