Chat model by Meta · 3.21B parameters
Llama 3.2 3B Instruct fits comfortably in your 24 GB Mac using Q8_0 quantization, using 17% of your RAM.
~50 tok/s
Q8_0 · 4.1 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 4.1 GB | ~50 tok/s | ✓ |
| Q6_K | 3.3 GB | ~66 tok/s | ✓ |
| Q5_K_M | 2.9 GB | ~77 tok/s | ✓ |
| Q4_K_M | 2.6 GB | ~91 tok/s | ✓ |
| Q3_K_M | 2.3 GB | ~110 tok/s | ✓ |
| Q2_K | 1.9 GB | ~142 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free