← Back to community benchmarks

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

M2 Max (30c) · 32 GB · 4bit · 2026-05-18
Performance
8k
tokens
154.4
PP tok/s
16.8
TG tok/s
53072
TTFT (ms)
18.9
Peak mem (GB)
Hardware
Chip M2 Max (30c)
Memory 32 GB
GPU Cores 30
Software
oMLX v0.3.9.dev2
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 152.4 20.0 16.4 GB view
4k 157.4 19.1 17.9 GB view
8k 154.4 16.8 18.9 GB current
Batching Results
Batch Size TG tok/s Speedup
20.0 1.00×
28.9 1.44×
57.9 2.90×
115.7 5.79×