← Back to community benchmarks

Qwen3.5-9B-OptiQ

M2 Pro (19c) · 16 GB · 4bit · 2026-03-12
Performance
1k
tokens
196.2
PP tok/s
33.1
TG tok/s
5218
TTFT (ms)
6.8
Peak mem (GB)
Hardware
Chip M2 Pro (19c)
Memory 16 GB
GPU Cores 19
Software
oMLX v0.2.9
macOS macOS 26.3.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 196.2 33.1 6.8 GB current
4k 205.3 31.5 7.0 GB view
8k 204.4 30.0 7.2 GB view
Batching Results
Batch Size TG tok/s Speedup
33.1 1.00×
45.8 1.38×
49.8 1.50×