← Back to community benchmarks

Qwen3.5-9B

M4 (10c) · 16 GB · 4bit · 2026-03-11
Performance
8k
tokens
197.2
PP tok/s
16.6
TG tok/s
41549
TTFT (ms)
7.0
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.7
macOS macOS 26.2
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 199.4 17.9 6.5 GB view
4k 199.7 17.4 6.8 GB view
8k 197.2 16.6 7.0 GB current
Batching Results
Batch Size TG tok/s Speedup
17.9 1.00×
34.3 1.92×
44.0 2.46×