← Back to community benchmarks

Qwen3.5-9B

M4 (8c) · 16 GB · 4bit · 2026-03-09
Performance
4k
tokens
168.7
PP tok/s
20.5
TG tok/s
24286
TTFT (ms)
6.8
Peak mem (GB)
Hardware
Chip M4 (8c)
Memory 16 GB
GPU Cores 8
Software
oMLX v0.2.6
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 170.4 21.6 6.5 GB view
4k 168.7 20.5 6.8 GB current
Batching Results
Batch Size TG tok/s Speedup
21.6 1.00×
24.8 1.15×
21.8 1.01×