General model by ibm-granite · 0.35B parameters
granite 4.0 350m fits comfortably in your 24 GB Mac using Q8_0 quantization, using 4% of your RAM.
~459 tok/s
Q8_0 · 0.9 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 0.9 GB | ~459 tok/s | ✓ |
| Q6_K | 0.8 GB | ~603 tok/s | ✓ |
| Q5_K_M | 0.8 GB | ~709 tok/s | ✓ |
| Q4_K_M | 0.7 GB | ~832 tok/s | ✓ |
| Q3_K_M | 0.7 GB | ~1,005 tok/s | ✓ |
| Q2_K | 0.7 GB | ~1,304 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free