← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M5 Max (40c) · 128 GB · 4bit · 2026-05-16
Performance
16k
tokens
803.6
PP tok/s
42.4
TG tok/s
20387
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 786.6 46.9 16.7 GB view
4k 891.9 47.4 18.2 GB view
8k 846.8 44.0 18.8 GB view
16k 803.6 42.4 20.0 GB current