NVIDIA ChatRTX Now Supports More Language Models

Becoming a leader with its market share in the artificial intelligence sector NVIDIA, also takes a share of their growth by producing chips such as the famous H100 for companies that want to develop in this field. In this way, NVIDIA, which always maintains its position at the top of the market, also shows its skills in its own artificial intelligence tools.

Among these tools, most of which come with RTX graphics cards for end users, there are applications that make things easier, such as Canvas and Broadcast. In addition, the company’s chat bot is Chat with RTX, or as it is now known. ChatRTX there is. With the updates, the chat bot in question is now more powerful than ever.

More language models are now supported.

This application, which runs locally on your computer, meaning it is not dependent on the cloud, already supported models such as Mistral and Llama 2. Now Google’s Gemma model, ChatGLM3 and OpenAI’s CLIP Even the model was added to the list. Of course, although the name ChatRTX However, to use this application, you need a GPU from at least the RTX 30 series and at least 8 GB of memory on this GPU.

OpenAI model will be useful for searching and interacting with your photos. Other models also have different functions and you choose which one you want to use.

When you run it, ChatRTX creates a local server on your computer. Your own documents that you add to this server turn into data and as a result, ChatRTX, is a powerful search tool that you can use personally, even answering questions about your own documents. In addition, thanks to Whisper integration, voice interaction is now possible.

When you want to download it, it has a size of 36GB. What are your thoughts about ChatRTX?