Chat model by Alibaba · 3.09B parameters
Qwen2.5 3B Instruct fits comfortably in your 24 GB Mac using Q8_0 quantization, using 16% of your RAM.
~52 tok/s
Q8_0 · 3.9 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 3.9 GB | ~52 tok/s | ✓ |
| Q6_K | 3.2 GB | ~68 tok/s | ✓ |
| Q5_K_M | 2.8 GB | ~80 tok/s | ✓ |
| Q4_K_M | 2.5 GB | ~94 tok/s | ✓ |
| Q3_K_M | 2.2 GB | ~114 tok/s | ✓ |
| Q2_K | 1.8 GB | ~148 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free