← Back to community benchmarks

Qwen3.5-27B

M4 Pro (20c) · 48 GB · 4bit · 2026-03-09
Performance
8k
tokens
124.3
PP tok/s
14.2
TG tok/s
65922
TTFT (ms)
17.4
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 125.6 15.2 16.6 GB view
4k 125.9 14.8 17.0 GB view
8k 124.3 14.2 17.4 GB current
16k 121.0 12.3 18.3 GB view
32k 114.0 9.7 20.1 GB view
Batching Results
Batch Size TG tok/s Speedup
15.2 1.00×
29.7 1.95×
33.2 2.18×