← Back to community benchmarks

Qwen3.5-9B-OptiQ

M4 (10c) · 32 GB · 4bit · 2026-03-11
Performance
16k
tokens
213.0
PP tok/s
17.4
TG tok/s
76931
TTFT (ms)
7.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.8
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 215.8 19.2 6.6 GB view
4k 222.2 19.1 6.9 GB view
8k 217.7 18.3 7.1 GB view
16k 213.0 17.4 7.6 GB current
32k 195.4 15.1 8.7 GB view
Batching Results
Batch Size TG tok/s Speedup
19.2 1.00×
37.8 1.97×
49.8 2.59×