Hello!
I began diving into the world of Large Language Models (LLMs) last year and have recently been given the chance to contribute to an LLM project at my workplace. At the moment, I’m exploring various libraries or frameworks, with HuggingFace and Langchain particularly catching my interest. I’ve come to understand that utilizing Hugging Face requires making calls to LLM APIs. However, my company has strict policies against using external LLM APIs due to concerns over potential data leaks, even prompts. This has led me to question whether working with Hugging Face’s library might inadvertently result in my queries or training data being sent to Hugging Face’s servers.
I’m equally eager to find out if employing Langchain’s library might involve any form of data transfer.
Thanks! 