Chat model by DeepSeek · 16.38B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
deepseek moe 16b chat fits comfortably in your 24 GB Mac using Q8_0 quantization, using 78% of your RAM.
— tok/s
Q8_0 · 18.8 GB · Benchmark needed
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 18.8 GB | ~71 tok/s | ✓ |
| Q6_K | 14.7 GB | ~93 tok/s | ✓ |
| Q5_K_M | 12.7 GB | ~109 tok/s | ✓ |
| Q4_K_M | 11.1 GB | ~128 tok/s | ✓ |
| Q3_K_M | 9.4 GB | ~154 tok/s | ✓ |
| Q2_K | 7.6 GB | ~200 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