← Back to community benchmarks

Qwen3.6-27B-oQ4-fp16-mtp

M2 Max (38c) · 96 GB · 4bit · 2026-05-15
Performance
4k
tokens
196.2
PP tok/s
27.8
TG tok/s
20873
TTFT (ms)
18.2
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 96 GB
GPU Cores 38
Software
oMLX v0.3.9.dev2
macOS macOS 26.4.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 189.6 29.4 16.8 GB view
4k 196.2 27.8 18.2 GB current
8k 195.2 27.4 19.2 GB view
16k 190.1 25.4 20.7 GB view
32k 179.4 23.6 23.7 GB view