← Back to community benchmarks

Qwen3.6-27B-UD

M5 Pro (20c) · 64 GB · 4bit · 2026-04-30
Performance
16k
tokens
438.8
PP tok/s
10.4
TG tok/s
37340
TTFT (ms)
28.5
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8rc1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 425.1 11.2 25.2 GB view
4k 465.0 11.0 26.6 GB view
8k 459.5 10.4 27.3 GB view
16k 438.8 10.4 28.5 GB current
Batching Results
Batch Size TG tok/s Speedup
11.2 1.00×
14.7 1.31×
24.9 2.22×