General model by jangq-ai · 31B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Gemma 4 31B it JANG_4M can run but will be tight at 85% of RAM. Consider a lower quantization or closing other apps.
~9 tok/s
Q4_K_M · 20.5 GB · Estimated
Alternative: Q3_K_M (17.4 GB, ~11 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 35.1 GB | ~5 tok/s | ✗ |
| Q6_K | 27.3 GB | ~7 tok/s | ✗ |
| Q5_K_M | 23.6 GB | ~8 tok/s | ✗ |
| Q4_K_M Recommended | 20.5 GB | ~9 tok/s | ✓ |
| Q3_K_M | 17.4 GB | ~11 tok/s | ✓ |
| Q2_K | 14.0 GB | ~15 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