Chat with RTX from NVIDIA. Private AI chatbot on your computer's disk.

Chat with RTX from NVIDIA is free, works fully locally on the user's computer and does not need an internet connection. And this means a much higher level of data privacy and significantly faster work.

Chat with RTX from NVIDIA. Private AI chatbot on your computer's disk.
00:00 00:00

Summary

  • NVIDIA has introduced a demo version of its new AI chatbot, Chat with RTX, which operates locally on Windows PCs and allows users to create personalized AI chatbots.
  • The chatbots are based on either the Mistral or Llama 2 large language models and use data from the computer they're installed on.
  • Users can ask the chatbot to scan files and notes for specific information, supporting various file formats and YouTube videos or playlists.
  • As Chat with RTX operates locally, results are delivered quickly and user data remains on the device, allowing for the processing of sensitive data without internet connectivity or sharing with external entities.
  • Chat with RTX is designed to work on modern computers equipped with NVIDIA GeForce RTX 30 series graphics cards or newer, with at least 8 GB of VRAM, the latest NVIDIA drivers, and the Windows 10 or 11 operating system.
  • The tool offers significant value for those who care about data privacy and want more control over chatbot technology, reducing the risk of information leakage.
  • Chat with RTX can be used not only by technology enthusiasts but also in the business sector, education, or research, where data security and information personalization are key.

NVIDIA has presented a demonstration version of its new AI chatbot - Chat with RTX. The tool operates entirely locally on personal computers with the Windows system and allows users to create their own, personalized AI chatbots.

They are based on one of two large language models (LLM) - Mistral or Llama 2 - and use data from the computer on which they were installed.

Thanks to this, users, instead of manually searching files and notes for specific information, can simply ask the chatbot, which will scan the indicated documents in a few seconds and provide an answer along with the context. Chat with RTX supports various file formats: .txt, .pdf, .doc/.docx, or .xml, as well as YouTube videos or playlists.

"Because Chat with RTX operates locally on Windows computers, results are delivered quickly, and user data remains on the device. Chat with RTX also allows for the processing of sensitive data without internet connectivity or the need to share them with external entities" - NVIDIA assures in a post on its blog.

Chat with RTX - hardware requirements

Chat with RTX is designed to work on modern computers equipped with NVIDIA GeForce RTX 30 series graphics cards or newer, which have at least 8 GB of VRAM (video random access memory). The tool also requires the latest NVIDIA drivers and the Windows 10 or 11 operating system.

Chat with RTX – higher level of privacy

Chat with RTX offers significant value for people who care about the privacy of their data and want to have more control over chatbot technology. Local data processing significantly reduces the risk of information leakage, which is particularly important in the context of subsequent privacy breaches by modern AI-based tools.

Therefore, the tool can find application not only among homegrown technology enthusiasts, but also in the business sector, education or research, where data security and information personalization are key. 

Generative AI. Will copyright infringement lawsuits bury its future?
Many indications suggest that the year 2024 will be crucial for the further development of generative AI models trained on publicly available materials. Subsequent courts will decide whether and to what extent this technology infringes copyright.