General model by Meta · 6.74B parameters
Llama 2 7b hf fits comfortably in your 24 GB Mac using Q8_0 quantization, using 33% of your RAM.
~24 tok/s
Q8_0 · 8.0 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 8.0 GB | ~24 tok/s | ✓ |
| Q6_K | 6.3 GB | ~31 tok/s | ✓ |
| Q5_K_M | 5.5 GB | ~37 tok/s | ✓ |
| Q4_K_M | 4.9 GB | ~43 tok/s | ✓ |
| Q3_K_M | 4.2 GB | ~52 tok/s | ✓ |
| Q2_K | 3.4 GB | ~68 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free