← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 24 GB · 4bit · 2026-03-10
Performance
4k
tokens
186.7
PP tok/s
18.2
TG tok/s
21935
TTFT (ms)
6.8
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
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 191.8 19.1 6.5 GB view
4k 186.7 18.2 6.8 GB current
8k 122.7 8.9 7.0 GB view
Batching Results
Batch Size TG tok/s Speedup
19.1 1.00×
16.4 0.86×
25.1 1.31×