Multimodal model by Alibaba · 31.07B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Qwen3 VL 30B A3B Instruct can run but will be tight at 86% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q4_K_M · 20.6 GB · Benchmark needed
Alternative: Q3_K_M (17.4 GB, ~104 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 35.2 GB | ~47 tok/s | ✗ |
| Q6_K | 27.4 GB | ~62 tok/s | ✗ |
| Q5_K_M | 23.7 GB | ~73 tok/s | ✗ |
| Q4_K_M Recommended | 20.6 GB | ~86 tok/s | ✓ |
| Q3_K_M | 17.4 GB | ~104 tok/s | ✓ |
| Q2_K | 14.0 GB | ~134 tok/s | ✓ |
Closest matches by use case and parameter count.
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free