General model by cyankiwi · 36.98B parameters
Qwen3.5 35B A3B AWQ 8bit fits comfortably in your 24 GB Mac using Q2_K quantization, using 69% of your RAM.
~155 tok/s
Q2_K · 16.6 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 41.8 GB | ~55 tok/s | ✗ |
| Q6_K | 32.5 GB | ~72 tok/s | ✗ |
| Q5_K_M | 28.1 GB | ~84 tok/s | ✗ |
| Q4_K_M | 24.4 GB | ~99 tok/s | ✗ |
| Q3_K_M | 20.7 GB | ~119 tok/s | ✗ |
| Q2_K Recommended | 16.6 GB | ~155 tok/s | ✓ |
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free