Chat model by Alibaba · 14.32B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Qwen1.5 MoE A2.7B Chat fits comfortably in your 24 GB Mac using Q8_0 quantization, using 69% of your RAM.
— tok/s
Q8_0 · 16.5 GB · Benchmark needed
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 16.5 GB | ~99 tok/s | ✓ |
| Q6_K | 12.9 GB | ~130 tok/s | ✓ |
| Q5_K_M | 11.2 GB | ~153 tok/s | ✓ |
| Q4_K_M | 9.7 GB | ~179 tok/s | ✓ |
| Q3_K_M | 8.3 GB | ~217 tok/s | ✓ |
| Q2_K | 6.7 GB | ~281 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