← Back to community benchmarks

Qwen3.5-9B

M4 Pro (16c) · 48 GB · 4bit · 2026-03-10
Performance
16k
tokens
315.9
PP tok/s
34.7
TG tok/s
51872
TTFT (ms)
7.6
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 48 GB
GPU Cores 16
Software
oMLX v0.2.6
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 344.2 45.6 6.7 GB view
4k 348.1 43.5 7.0 GB view
8k 341.5 41.3 7.1 GB view
16k 315.9 34.7 7.6 GB current
Batching Results
Batch Size TG tok/s Speedup
45.6 1.00×
75.0 1.64×
80.8 1.77×