General model by huihui-ai · 35.95B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Huihui Qwen3.6 35B A3B abliterated can run but will be tight at 84% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q3_K_M · 20.1 GB · Benchmark needed
Alternative: Q2_K (16.2 GB, ~152 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 40.6 GB | ~54 tok/s | ✗ |
| Q6_K | 31.6 GB | ~70 tok/s | ✗ |
| Q5_K_M | 27.3 GB | ~83 tok/s | ✗ |
| Q4_K_M | 23.7 GB | ~97 tok/s | ✗ |
| Q3_K_M Recommended | 20.1 GB | ~117 tok/s | ✓ |
| Q2_K | 16.2 GB | ~152 tok/s | ✓ |
GGUF Sources
Closest matches by use case and parameter count.
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free