General model by languagebind · 1.6B parameters
Everything this chip runs, machine by machine: MacBook Pro 14" M5 Pro, MacBook Pro 16" M5 Pro
MoE LLaVA StableLM 1.6B 4e fits comfortably in your 24 GB Mac using Q8_0 quantization, using 10% of your RAM.
~101 tok/s
Q8_0 · 2.3 GB · Estimated
| Quantization | Memory | Speed | Fits? |
|---|---|---|---|
| Q8_0 Recommended | 2.3 GB | ~101 tok/s | ✓ |
| Q6_K | 1.9 GB | ~132 tok/s | ✓ |
| Q5_K_M | 1.7 GB | ~155 tok/s | ✓ |
| Q4_K_M | 1.5 GB | ~182 tok/s | ✓ |
| Q3_K_M | 1.4 GB | ~220 tok/s | ✓ |
| Q2_K | 1.2 GB | ~285 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