← Back to community benchmarks

Qwen3.5-9B

M1 (7c) · 16 GB · 4bit · 2026-03-13
Performance
1k
tokens
69.4
PP tok/s
12.7
TG tok/s
14748
TTFT (ms)
6.5
Peak mem (GB)
Hardware
Chip M1 (7c)
Memory 16 GB
GPU Cores 7
Software
oMLX v0.2.9
macOS macOS 26.2
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 69.4 12.7 6.5 GB current
4k 69.6 12.6 6.8 GB view
Batching Results
Batch Size TG tok/s Speedup
12.7 1.00×
16.6 1.31×
15.9 1.25×
14.3 1.13×