Multimodal model by davidau · 39.53B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Qwen3.5 40B RoughHouse Claude 4.6 Opus Polar Deckard Uncensored Heretic Thinking can run but will be tight at 74% of RAM. Consider a lower quantization or closing other apps.
~12 tok/s
Q2_K · 17.7 GB · Estimated
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 44.6 GB | ~4 tok/s | ✗ |
| Q6_K | 34.7 GB | ~5 tok/s | ✗ |
| Q5_K_M | 30.0 GB | ~6 tok/s | ✗ |
| Q4_K_M | 26.0 GB | ~7 tok/s | ✗ |
| Q3_K_M | 22.1 GB | ~9 tok/s | ✗ |
| Q2_K Recommended | 17.7 GB | ~12 tok/s | ✓ |
Affiliate disclosure: the shop icons in this table are Amazon affiliate links. As an Amazon Associate, ModelPiper earns from qualifying purchases. Every fit rating, quantization and tok/s figure below is computed from the hardware specs and is not influenced by that.
Closest matches by use case and parameter count.
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper free