RAG stands for Retrieval Augmented Generation. RAG is a technique that enables large language models (LLMs) to retrieve and incorporate new information. When using a public LLM, it has not been trained with your company's specific data, unless it's public data. Using RAG with GhostAI allows an administrator to add company specific data, which then can be referenced during prompts.
To enable RAG, click "Choose Files" to select which documents you want to upload. Then next to the file name, select which groups have access to the information contained in this file. You can also select groups in the "Bulk Assign Groups to All Files" in case you want specific groups to have access to all information. After selecting the groups, click "Upload". After uploading the file(s), the "RAG Search" toggle will now be active and you can set it to "Enabled".
Note that processing the files might take a while, and will be done in the background. During this processing, it is not yet possible to ask questions about the data using GhostAI