The realm of ‘chatbots’ has been significantly expanded with a fresh proposal from Nvidia. This technology giant has launched a tool that allows users to customize a conversational assistant. The assistant is powered by generative artificial intelligence (AI) and uniquely, it operates locally on a computer running Windows, provided that it is equipped with GeForce RTX 30 Series graphics.
The tool, aptly named Chat with RTX, facilitates the application of generative AI in a local environment. This allows users to tailor the content of a ‘chatbot’ as per their requirements. It is a notable advancement in AI technology, bringing the capabilities of AI closer to the users.
Chat with RTX employs the Recall Augmented Generation (RAG) technique. This technique improves the accuracy and reliability of AI models. In addition, it utilizes Nvidia’s TensorRT-LLM software and NVIDIA RTX acceleration to further enhance its performance.
The tool is designed to enable the creation of a personalized ‘chatbot’ that can handle queries in a conversational format. Similar to chatbots like Copilot, Gemini or ChatGPT, it supports different text formats. An interesting feature of the tool is that it allows the incorporation of YouTube video URLs for contextual queries.
The ‘chatbot’ is suitable for use on Windows 10 or 11 computers that meet certain minimum specifications. These include a GPU GeForce RTX from the 30 Series and 8GB of video random access memory (VRAM).
One of the key advantages of local execution is the enhanced privacy it offers. Since it does not rely on cloud services, users can process sensitive data on a local PC. This eliminates the need to share data with a third party or have an Internet connection, as Nvidia explains in their news blog.
