← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 24 GB · 4bit · 2026-03-11
Performance
16k
tokens
205.8
PP tok/s
17.9
TG tok/s
79617
TTFT (ms)
7.5
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 26.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 216.5 20.9 6.5 GB view
4k 213.0 18.4 6.8 GB view
8k 216.2 19.6 7.0 GB view
16k 205.8 17.9 7.5 GB current
32k 182.5 15.3 8.5 GB view
64k 157.9 12.0 10.6 GB view
Batching Results
Batch Size TG tok/s Speedup
20.9 1.00×
40.7 1.95×
49.5 2.37×