Chat model by redhatai · 1.5B parameters
Llama 3.2 1B Instruct FP8 dynamic fits comfortably in your 24 GB Mac using Q8_0 quantization, using 9% of your RAM.
~107 tok/s
Q8_0 · 2.2 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 2.2 GB | ~107 tok/s | ✓ |
| Q6_K | 1.8 GB | ~141 tok/s | ✓ |
| Q5_K_M | 1.6 GB | ~166 tok/s | ✓ |
| Q4_K_M | 1.5 GB | ~194 tok/s | ✓ |
| Q3_K_M | 1.3 GB | ~235 tok/s | ✓ |
| Q2_K | 1.2 GB | ~304 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free