General model by opengvlab · 2.21B parameters
InternVL2 2B fits comfortably in your 24 GB Mac using Q8_0 quantization, using 12% of your RAM.
~73 tok/s
Q8_0 · 3.0 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 3.0 GB | ~73 tok/s | ✓ |
| Q6_K | 2.4 GB | ~96 tok/s | ✓ |
| Q5_K_M | 2.1 GB | ~112 tok/s | ✓ |
| Q4_K_M | 1.9 GB | ~132 tok/s | ✓ |
| Q3_K_M | 1.7 GB | ~159 tok/s | ✓ |
| Q2_K | 1.5 GB | ~206 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free