← Back to community benchmarks

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

M5 Pro (20c) · 48 GB · 4bit · 2026-05-23
Performance
16k
tokens
447.3
PP tok/s
23.9
TG tok/s
36626
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.9
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 433.8 26.3 16.7 GB view
4k 474.4 26.5 18.2 GB view
8k 469.7 25.4 19.2 GB view
16k 447.3 23.9 20.7 GB current
32k 407.0 21.8 23.7 GB view
64k 339.6 18.2 29.7 GB view
Batching Results
Batch Size TG tok/s Speedup
26.3 1.00×
31.6 1.20×
41.9 1.59×
44.9 1.71×