← Back to community benchmarks

Qwen3.6-27B-oQ8-mtp

M5 Max (40c) · 128 GB · 8bit · 2026-05-09
Performance
64k
tokens
508.0
PP tok/s
20.9
TG tok/s
128996
TTFT (ms)
41.8
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 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 701.9 29.0 28.8 GB view
4k 802.5 28.6 30.3 GB view
8k 754.4 26.4 31.3 GB view
16k 657.7 23.9 32.8 GB view
32k 600.5 24.2 35.8 GB view
64k 508.0 20.9 41.8 GB current
128k 384.1 17.9 54.2 GB view