← Back to community benchmarks

Qwen3.5-9B

M4 Pro (16c) · 24 GB · 4bit · 2026-03-10
Performance
4k
tokens
121.6
PP tok/s
17.9
TG tok/s
33698
TTFT (ms)
7.0
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 24 GB
GPU Cores 16
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 176.3 22.3 6.7 GB view
4k 121.6 17.9 7.0 GB current
8k 162.0 26.5 7.1 GB view
16k 204.9 23.0 7.6 GB view
32k 138.2 12.2 8.7 GB view
Batching Results
Batch Size TG tok/s Speedup
22.3 1.00×
11.7 0.52×
14.3 0.64×
13.5 0.61×