← Back to community benchmarks

Qwen3.5-9B

M5 Pro (20c) · 64 GB · 4bit · 2026-07-06
Performance
128k
tokens
755.8
PP tok/s
26.4
TG tok/s
173421
TTFT (ms)
19.6
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.4.4
macOS macOS 26.5.2
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,258 60.0 5.8 GB view
8k 1,591 56.5 6.9 GB view
32k 1,384 47.8 9.4 GB view
128k 755.8 26.4 19.6 GB current