← Back to community benchmarks

Qwen3.6-27B

M5 Max (40c) · 128 GB · 4bit · 2026-06-17
Performance
195k
tokens
302.6
PP tok/s
16.6
TG tok/s
660975
TTFT (ms)
47.8
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.4.4
macOS macOS 26.5.1
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 790.9 33.9 15.9 GB view
4k 928.6 32.8 17.3 GB view
8k 849.2 33.0 17.9 GB view
16k 775.5 31.5 19.2 GB view
32k 701.9 29.6 21.7 GB view
64k 562.8 25.2 26.7 GB view
128k 393.8 19.9 37.1 GB view
195k 302.6 16.6 47.8 GB current
Batching Results
Batch Size TG tok/s Speedup
33.9 1.00×
56.8 1.68×
61.4 1.81×