← Back to community benchmarks

Qwen3.5-VL-9B-4bitCRACK

M4 (10c) · 16 GB · 4bit · 2026-03-13
Performance
64k
tokens
163.8
PP tok/s
8.6
TG tok/s
400070
TTFT (ms)
10.6
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.2.9
macOS macOS 26.2
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 225.2 21.9 6.5 GB view
4k 226.1 21.2 6.8 GB view
8k 222.0 19.3 7.0 GB view
16k 202.9 16.5 7.5 GB view
32k 187.5 12.4 8.5 GB view
64k 163.8 8.6 10.6 GB current
Batching Results
Batch Size TG tok/s Speedup
21.9 1.00×
42.3 1.93×
50.1 2.29×
50.4 2.30×