← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 32 GB · 4bit · 2026-03-11
Performance
16k
tokens
214.5
PP tok/s
18.7
TG tok/s
76393
TTFT (ms)
7.5
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 220.2 21.1 6.5 GB view
4k 225.1 21.1 6.8 GB view
8k 222.2 20.3 7.0 GB view
16k 214.5 18.7 7.5 GB current
32k 196.1 15.9 8.5 GB view
Batching Results
Batch Size TG tok/s Speedup
21.1 1.00×
42.7 2.02×