With logging being DEBUG, i see that txtai send requests every time i initialize LLM() class.

DEBUG 2024-01-09 23:58:56,704 Starting new HTTPS connection (1): huggingface.co:443
DEBUG 2024-01-09 23:58:57,075 https://huggingface.co:443 "HEAD /TheBloke/TinyLlama-1.1B-Chat-v1.0-GGUF/resolve/main/tinyllama-1.1b-chat-v1.0.Q4_K_M.gguf HTTP/1.1" 302 0

I have TinyLlama downloaded and would not like to block a service for a short while by them.

Actually i have two questions here:

1. How to disable this HEAD request?
2. How to manually download a model, should i do it without txtai or with?