← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 24 GB · 4bit · 2026-03-12
Performance
32k
tokens
102.2
PP tok/s
8.3
TG tok/s
320531
TTFT (ms)
8.5
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.2.9
macOS macOS 26.3.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 177.5 18.6 6.5 GB view
4k 171.4 16.4 6.8 GB view
8k 151.8 14.4 7.0 GB view
16k 85.3 11.0 7.5 GB view
32k 102.2 8.3 8.5 GB current
Batching Results
Batch Size TG tok/s Speedup
18.6 1.00×
21.1 1.13×
25.8 1.39×
26.1 1.40×