General model by z-lab · 26B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
gemma 4 26B A4B it DFlash can run but will be tight at 83% of RAM. Consider a lower quantization or closing other apps.
~10 tok/s
Q5_K_M · 19.9 GB · Estimated
Alternative: Q4_K_M (17.3 GB, ~11 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 29.5 GB | ~6 tok/s | ✗ |
| Q6_K | 23.0 GB | ~8 tok/s | ✗ |
| Q5_K_M Recommended | 19.9 GB | ~10 tok/s | ✓ |
| Q4_K_M | 17.3 GB | ~11 tok/s | ✓ |
| Q3_K_M | 14.7 GB | ~14 tok/s | ✓ |
| Q2_K | 11.8 GB | ~18 tok/s | ✓ |
Closest matches by use case and parameter count.
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free