General model by alibaba-nlp · 30.53B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
Tongyi DeepResearch 30B A3B can run but will be tight at 84% of RAM. Consider a lower quantization or closing other apps.
— tok/s
Q4_K_M · 20.2 GB · Benchmark needed
Alternative: Q3_K_M (17.2 GB, ~105 tok/s)
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 | 34.6 GB | ~48 tok/s | ✗ |
| Q6_K | 26.9 GB | ~63 tok/s | ✗ |
| Q5_K_M | 23.3 GB | ~74 tok/s | ✗ |
| Q4_K_M Recommended | 20.2 GB | ~87 tok/s | ✓ |
| Q3_K_M | 17.2 GB | ~105 tok/s | ✓ |
| Q2_K | 13.8 GB | ~137 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