Chat model by Microsoft · 41.87B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Phi 3.5 MoE instruct can run but will be tight at 78% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q2_K · 18.7 GB · Benchmark needed
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 47.2 GB | ~23 tok/s | ✗ |
| Q6_K | 36.7 GB | ~30 tok/s | ✗ |
| Q5_K_M | 31.7 GB | ~35 tok/s | ✗ |
| Q4_K_M | 27.5 GB | ~41 tok/s | ✗ |
| Q3_K_M | 23.3 GB | ~50 tok/s | ✗ |
| Q2_K Recommended | 18.7 GB | ~65 tok/s | ✓ |
GGUF Sources
Closest matches by use case and parameter count.
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free