← Back to community benchmarks

Qwen3.5-9B-OptiQ

M2 Pro (16c) · 16 GB · 4bit · 2026-03-15
Performance
8k
tokens
173.4
PP tok/s
30.0
TG tok/s
47232
TTFT (ms)
7.2
Peak mem (GB)
Hardware
Chip M2 Pro (16c)
Memory 16 GB
GPU Cores 16
Software
oMLX v0.2.12
macOS macOS 26.3.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 169.0 33.0 6.8 GB view
4k 174.5 31.7 7.0 GB view
8k 173.4 30.0 7.2 GB current
Batching Results
Batch Size TG tok/s Speedup
33.0 1.00×
40.1 1.22×
42.9 1.30×