meta-llama/llama3 ? reverse-engineered prompt

Hey, can you help me get Meta's Llama 3 model running on my computer? I found what looks like the official GitHub repo, and I want to set it up so I can try out the example scripts they provide for chat completion and basic text generation.

It looks like there's a process for downloading the models from Meta or Hugging Face that seems a bit complicated. Can you please handle setting up the toàn bộ project? Install all the Python dependencies for a machine with a CUDA GPU. Please guide me through getting the 8B instruction-tuned model specifically. Once all the setup and downloads are done, just show me the simple terminal command I'd use to actually run the chat example.