← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M5 Max (40c) · 128 GB · 8bit · 2026-05-23
Performance
8k
tokens
805.9
PP tok/s
28.7
TG tok/s
10165
TTFT (ms)
31.3
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 729.6 29.7 28.8 GB view
4k 851.4 29.2 30.3 GB view
8k 805.9 28.7 31.3 GB current
16k 711.5 27.5 32.8 GB view
32k 621.7 24.4 35.8 GB view
64k 534.2 21.0 41.8 GB view