← Back to community benchmarks

Qwen3.5-9B

M5 (10c) · 24 GB · 4bit · 2026-04-15
Performance
16k
tokens
640.9
PP tok/s
23.3
TG tok/s
25564
TTFT (ms)
7.4
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.5
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
4k 684.0 25.5 6.3 GB view
8k 678.1 24.6 6.7 GB view
16k 640.9 23.3 7.4 GB current
32k 542.9 20.1 8.9 GB view