← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 32 GB · 4bit · 2026-03-21
Performance
4k
tokens
187.6
PP tok/s
18.5
TG tok/s
21829
TTFT (ms)
7.3
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.2.19
macOS macOS 15.6.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 194.8 18.9 6.5 GB view
4k 187.6 18.5 7.3 GB current
Batching Results
Batch Size TG tok/s Speedup
1× 18.9 1.00×
2× 37.1 1.96×
4× 42.0 2.22×
8× 39.2 2.07×