General model by opengvlab · 9.14B parameters
InternVL3 9B fits comfortably in your 24 GB Mac using Q8_0 quantization, using 45% of your RAM.
~18 tok/s
Q8_0 · 10.7 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 10.7 GB | ~18 tok/s | ✓ |
| Q6_K | 8.4 GB | ~23 tok/s | ✓ |
| Q5_K_M | 7.3 GB | ~27 tok/s | ✓ |
| Q4_K_M | 6.4 GB | ~32 tok/s | ✓ |
| Q3_K_M | 5.5 GB | ~38 tok/s | ✓ |
| Q2_K | 4.5 GB | ~50 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free