← Back to community benchmarks

Qwen3.6-27B-MTPLX-Optimized

M5 Max (40c) · 128 GB · 4bit · 2026-05-19
Performance
8k
tokens
857.9
PP tok/s
23.9
TG tok/s
9549
TTFT (ms)
22.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9rc1
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 780.4 24.5 20.0 GB view
4k 900.9 23.9 21.5 GB view
8k 857.9 23.9 22.5 GB current
16k 814.6 23.2 24.0 GB view
32k 726.4 21.8 27.0 GB view
64k 614.1 19.9 33.0 GB view