Chat model by Mistral AI · 7.24B parameters
Mistral 7B Instruct v0.2 fits comfortably in your 24 GB Mac using Q8_0 quantization, using 36% of your RAM.
~22 tok/s
Q8_0 · 8.6 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 8.6 GB | ~22 tok/s | ✓ |
| Q6_K | 6.8 GB | ~29 tok/s | ✓ |
| Q5_K_M | 5.9 GB | ~34 tok/s | ✓ |
| Q4_K_M | 5.2 GB | ~40 tok/s | ✓ |
| Q3_K_M | 4.4 GB | ~49 tok/s | ✓ |
| Q2_K | 3.7 GB | ~63 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free