← Back to community benchmarks

Qwen3.5-9B-OptiQ

M5 (10c) · 16 GB · 4bit · 2026-03-19
Performance
4k
tokens
722.3
PP tok/s
22.8
TG tok/s
5671
TTFT (ms)
7.4
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.19
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 681.3 22.6 6.7 GB view
4k 722.3 22.8 7.4 GB current
Batching Results
Batch Size TG tok/s Speedup
22.6 1.00×
48.4 2.14×
68.9 3.05×