← Back to community benchmarks

Qwen3.5-27B

M4 Pro (20c) · 64 GB · 4bit · 2026-03-13
Performance
16k
tokens
115.1
PP tok/s
12.9
TG tok/s
142335
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.6
macOS macOS 15.6.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 126.5 15.3 16.6 GB view
4k 125.4 14.7 17.0 GB view
8k 115.2 14.1 17.4 GB view
16k 115.1 12.9 18.3 GB current
Batching Results
Batch Size TG tok/s Speedup
15.3 1.00×
29.0 1.90×
33.1 2.16×
33.8 2.21×