← Back to community benchmarks

qwen3.6-dense-27b-mtp

M5 Max (40c) · 128 GB · 4bit · 2026-05-10
Performance
4k
tokens
720.3
PP tok/s
40.7
TG tok/s
5687
TTFT (ms)
18.2
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 668.9 42.9 16.8 GB view
4k 720.3 40.7 18.2 GB current
8k 682.6 38.5 19.2 GB view
Batching Results
Batch Size TG tok/s Speedup
42.9 1.00×
49.1 1.14×
58.8 1.37×
52.4 1.22×