General model by nvidia · 31.58B parameters
NVIDIA Nemotron 3 Nano 30B A3B FP8 can run but will be tight at 74% of RAM. Consider a lower quantization or closing other apps.
~11 tok/s
Q3_K_M · 17.7 GB
Alternative: Q2_K (14.3 GB, ~14 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 35.7 GB | ~5 tok/s | ✗ |
| Q6_K | 27.8 GB | ~7 tok/s | ✗ |
| Q5_K_M | 24.0 GB | ~8 tok/s | ✗ |
| Q4_K_M | 20.9 GB | ~9 tok/s | ✗ |
| Q3_K_M Recommended | 17.7 GB | ~11 tok/s | ✓ |
| Q2_K | 14.3 GB | ~14 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free