← Back to community benchmarks

Qwen3.5-9B

M4 · 12 GB · 4bit · 2026-03-15
Performance
4k
tokens
206.7
PP tok/s
18.2
TG tok/s
19821
TTFT (ms)
6.8
Peak mem (GB)
Hardware
Chip M4
Memory 12 GB
Software
oMLX v0.2.12
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 190.6 17.8 6.5 GB view
4k 206.7 18.2 6.8 GB current
Batching Results
Batch Size TG tok/s Speedup
17.8 1.00×
35.9 2.02×
51.0 2.87×