General model by trevorjs · 25.81B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
gemma 4 26B A4B it uncensored can run but will be tight at 82% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q5_K_M · 19.7 GB · Benchmark needed
Alternative: Q4_K_M (17.2 GB, ~103 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 29.3 GB | ~57 tok/s | ✗ |
| Q6_K | 22.8 GB | ~75 tok/s | ✗ |
| Q5_K_M Recommended | 19.7 GB | ~88 tok/s | ✓ |
| Q4_K_M | 17.2 GB | ~103 tok/s | ✓ |
| Q3_K_M | 14.6 GB | ~125 tok/s | ✓ |
| Q2_K | 11.7 GB | ~162 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