← Back to community benchmarks

Qwen3.5-9B-OptiQ

M4 (10c) · 24 GB · 4bit · 2026-04-30
Performance
16k
tokens
198.4
PP tok/s
17.1
TG tok/s
82586
TTFT (ms)
8.4
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.7
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 205.9 18.7 6.7 GB view
4k 220.8 19.1 7.3 GB view
8k 212.2 18.3 7.7 GB view
16k 198.4 17.1 8.4 GB current
Batching Results
Batch Size TG tok/s Speedup
18.7 1.00×
38.2 2.04×
46.8 2.50×