← Back to community benchmarks

Qwen3.5-9B

M1 Max (24c) · 32 GB · 8bit · 2026-03-09
Performance
64k
tokens
205.0
PP tok/s
13.9
TG tok/s
319694
TTFT (ms)
14.8
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 32 GB
GPU Cores 24
Software
oMLX v0.2.6
macOS macOS 26.3
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 239.3 35.1 10.7 GB view
4k 242.1 33.3 11.0 GB view
8k 236.2 30.3 11.2 GB view
16k 234.4 26.2 11.7 GB view
32k 224.8 20.3 12.7 GB view
64k 205.0 13.9 14.8 GB current
Batching Results
Batch Size TG tok/s Speedup
35.1 1.00×
53.4 1.52×
58.4 1.66×
46.4 1.32×