← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M5 Max (40c) · 128 GB · 8bit · 2026-06-14
User commentDefault parameters
Performance
128k
tokens
304.8
PP tok/s
10.9
TG tok/s
429979
TTFT (ms)
53.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.4.4rc1
macOS macOS 26.5.1
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 722.1 18.8 28.3 GB view
4k 839.4 18.6 29.8 GB view
8k 803.7 18.3 30.8 GB view
16k 686.7 17.7 32.3 GB view
32k 618.1 17.2 35.3 GB view
64k 439.7 14.6 41.4 GB view
128k 304.8 10.9 53.7 GB current
195k 222.7 9.8 68.4 GB view