General model by dealignai · 31.37B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
DeepSeek V4 Flash 0731 JANG CRACK can run but will be tight at 73% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q3_K_M · 17.6 GB · Benchmark needed
Alternative: Q2_K (14.2 GB, ~201 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 35.5 GB | ~71 tok/s | ✗ |
| Q6_K | 27.7 GB | ~93 tok/s | ✗ |
| Q5_K_M | 23.9 GB | ~110 tok/s | ✗ |
| Q4_K_M | 20.8 GB | ~128 tok/s | ✗ |
| Q3_K_M Recommended | 17.6 GB | ~155 tok/s | ✓ |
| Q2_K | 14.2 GB | ~201 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