← Back to community benchmarks

Qwen3.5-9B

M1 Pro (16c) · 16 GB · 4bit · 2026-03-10
Performance
8k
tokens
157.0
PP tok/s
29.2
TG tok/s
52178
TTFT (ms)
7.1
Peak mem (GB)
Hardware
Chip M1 Pro (16c)
Memory 16 GB
GPU Cores 16
Software
oMLX v0.2.6
macOS macOS 26.3
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 157.9 28.6 6.7 GB view
4k 152.6 30.8 7.0 GB view
8k 157.0 29.2 7.1 GB current
Batching Results
Batch Size TG tok/s Speedup
28.6 1.00×
36.2 1.27×
36.4 1.27×
31.4 1.10×