← Back to community benchmarks

Qwen3.6-27B-oQ4-mtp

M5 Pro (20c) · 48 GB · 4bit · 2026-05-20
Performance
32k
tokens
410.6
PP tok/s
21.8
TG tok/s
79799
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.9rc1
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 423.2 23.9 16.8 GB view
4k 467.5 25.6 18.2 GB view
8k 466.0 24.8 19.2 GB view
16k 443.9 24.0 20.7 GB view
32k 410.6 21.8 23.7 GB current
64k 339.7 17.7 29.7 GB view
Batching Results
Batch Size TG tok/s Speedup
23.9 1.00×
48.2 2.02×
104.2 4.36×