Skip to main content

Intro to LLM

By now you've probably heard of Large Language Models (LLMs). They're the AI models that are currently powering ChatGPT, Google's Bard, and other chatbots.

PrivateGPT uses LLMs in combination with RAG and other optimzation techniques to answer questions.

While LLMs are great for general purpose question answering, they're not always able to answer your questions correctly. And although they have been trained on a lot of public data, they do not have access to your organization's private data. Furthermore, they are not able to answer questions about events that happened after they finished training. Another limitation is that they tend to "hallucinate", meaning they sometimes make up information that is not correct.

PrivateGPT addresses these limitations by using LLMs in such a way that they have secure access to your organization's private data. PrivateGPT comes with fine-grained access control, so that you can control which users have access to which data.

Next to that, PrivateGPT comes with a set of models that are optimized for specific industries and use cases. These models are trained on industry-specific data and are able to answer questions about that data, especially when combined with RAG.

Models | "Supercharge Your Private GPT with Customized Expertise"​

Unlock Peak Performance by integrating Specialized Skills (Models) into your Private GPT via our streamlined Best-Practice flow, tailored to your unique application needs. Choose from:

  • Open-Source Models: Leverage the power of community-driven innovation
  • Industry-Specific Models: Benefit from expertly crafted, sector-tailored solutions designed to address your industry's distinct challenges

Available models:

  • LLM | Legal
  • LLM | Accountancy
  • LLM | Code
  • LLM | Healthcare
  • LLM | Local Government
  • And many more, soon…