General model by poolside · 33.44B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Laguna XS.2 can run but will be tight at 78% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q3_K_M · 18.7 GB · Benchmark needed
Alternative: Q2_K (15.1 GB, ~171 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 37.8 GB | ~60 tok/s | ✗ |
| Q6_K | 29.4 GB | ~79 tok/s | ✗ |
| Q5_K_M | 25.4 GB | ~93 tok/s | ✗ |
| Q4_K_M | 22.1 GB | ~109 tok/s | ✗ |
| Q3_K_M Recommended | 18.7 GB | ~132 tok/s | ✓ |
| Q2_K | 15.1 GB | ~171 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