← Back to community benchmarks

community:Qwen3.5-9B-OptiQ

M4 (10c) · 32 GB · 4bit · 2026-03-11
Performance
16k
tokens
203.5
PP tok/s
17.1
TG tok/s
80506
TTFT (ms)
7.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 212.8 19.3 6.6 GB view
4k 221.9 19.1 6.9 GB view
8k 219.9 18.6 7.1 GB view
16k 203.5 17.1 7.6 GB current
32k 186.3 14.7 8.7 GB view
Batching Results
Batch Size TG tok/s Speedup
19.3 1.00×
38.1 1.97×