General model by davidau · 40.65B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
gemma 3 it vl 40B Gemini Heretic Uncensored Thinking can run but will be tight at 76% of RAM. Consider a lower quantization or closing other apps.
~11 tok/s
Q2_K · 18.2 GB · Estimated
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 45.8 GB | ~4 tok/s | ✗ |
| Q6_K | 35.7 GB | ~5 tok/s | ✗ |
| Q5_K_M | 30.8 GB | ~6 tok/s | ✗ |
| Q4_K_M | 26.7 GB | ~7 tok/s | ✗ |
| Q3_K_M | 22.7 GB | ~9 tok/s | ✗ |
| Q2_K Recommended | 18.2 GB | ~11 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