General model by 0xsero · 28.24B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Qwen3.6 28B REAP can run but will be tight at 78% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q4_K_M · 18.7 GB · Benchmark needed
Alternative: Q3_K_M (15.9 GB, ~143 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 32.0 GB | ~65 tok/s | ✗ |
| Q6_K | 24.9 GB | ~86 tok/s | ✗ |
| Q5_K_M | 21.6 GB | ~101 tok/s | ✗ |
| Q4_K_M Recommended | 18.7 GB | ~118 tok/s | ✓ |
| Q3_K_M | 15.9 GB | ~143 tok/s | ✓ |
| Q2_K | 12.8 GB | ~186 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