Skip to main content

What is an LLM, and what is RAG in AskTuring?

Describes what an LLM is, how RAG works in AskTuring, and how your documents interact with the AI model.

Eyal Leeder avatar
Written by Eyal Leeder
Updated over 3 weeks ago

Short answer:


An LLM (Large Language Model) is the AI “brain” that understands and generates text. RAG (Retrieval-Augmented Generation) is the process AskTuring uses to ground that intelligence in your documents, ensuring that answers are accurate, relevant, and specific to your data.

Full explanation:

In AskTuring, the Large Language Model (LLM) is the engine that interprets your question, understands its meaning, and creates a response.


You can choose which LLM to use for your chats, allowing you to select the one that best fits your needs for tone, speed, accuracy, or depth of reasoning. You can even switch between different LLMs within the same conversation.

What is RAG (Retrieval-Augmented Generation)?

RAG is the method AskTuring uses to ensure that the AI’s responses are grounded in your actual documents rather than relying only on general knowledge.

Here’s how it works:

  1. When you ask a question, AskTuring searches your selected Project (or chosen folders and files) for the most relevant information.

  2. That information is retrieved and passed to the LLM.

  3. The LLM uses it to craft a response that’s specific to your content and context.

This process keeps answers factual, verifiable, and directly linked to your uploaded materials.

Why It Matters

RAG keeps responses grounded in the right source, your verified Project documents, while still allowing the LLM to express the information clearly and naturally.
It’s what elevates AskTuring from a general-purpose AI to a specialized expert assistant that works precisely with your organization’s data, ensuring:

  • Transparency through document citations and references

  • Accuracy from verified internal sources

  • Flexibility to adapt its tone and depth based on your chosen LLM

Tips

  • Choose the LLM that best fits your task; some models excel at summarization, others at analysis or writing.

  • Review the document citations in responses to confirm exactly where the information came from.

  • Keep your Projects updated so the LLM always has access to your most recent and accurate content.

  • Remember that AskTuring never allows any LLM to retain, store, or train on your private data.

Did this answer your question?