General model by openai · 21.51B parameters
gpt oss 20b fits comfortably in your 24 GB Mac using Q6_K quantization, using 80% of your RAM.
~58 tok/s
Q6_K · 19.1 GB
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 24.5 GB | ~44 tok/s | ✗ |
| Q6_K Recommended | 19.1 GB | ~58 tok/s | ✓ |
| Q5_K_M | 16.5 GB | ~68 tok/s | ✓ |
| Q4_K_M | 14.4 GB | ~80 tok/s | ✓ |
| Q3_K_M | 12.2 GB | ~97 tok/s | ✓ |
| Q2_K | 9.9 GB | ~126 tok/s | ✓ |
GGUF Sources
ToolPiper downloads, manages, and runs models with one click. Apple Silicon optimized.
Get ToolPiper — Free