Chat model by ai-sage · 11.48B parameters
GigaChat3 10B A1.8B fits comfortably in your 24 GB Mac using Q8_0 quantization, using 55% of your RAM.
~216 tok/s
Q8_0 · 13.3 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 13.3 GB | ~216 tok/s | ✓ |
| Q6_K | 10.4 GB | ~284 tok/s | ✓ |
| Q5_K_M | 9.1 GB | ~334 tok/s | ✓ |
| Q4_K_M | 7.9 GB | ~391 tok/s | ✓ |
| Q3_K_M | 6.8 GB | ~473 tok/s | ✓ |
| Q2_K | 5.5 GB | ~613 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free