NVIDIA launches free AI chatbot 'Chat With RTX', which can run natively on a PC with a GeForce RTX GPU

NVIDIA technology demo running on GeForce RTX GPUsChat with RTXIt is released for free. Users can customize chat with RTX with their own content, and system requirements are a Windows machine with an RTX 30 series or newer GPU with at least 8GB of VRAM.

Chat with RTX now for free to download | Nvidia blog
https://blogs.nvidia.com/blog/chat-with-rtx-available-now/

NVIDIA brings generative AI to millions, with Tensor Core GPUs, LLMs, and tools for PCs and RTX workstations | Nvidia Newsroom
https://nvidianews.nvidia.com/news/geneative-ai-rtx-pcs-and-workstations

Chat with RTXRecovery Augmented Generation (RAG)、TensorRT-LLMRun locally generated AI on your Windows PC with NVIDIA RTX acceleration technology. Users can use local files on their computers as datasets.MistralorLlama 2It is said that it can be quickly and easily plugged into large-scale open source language models such as .

You can see how chat with RTX works in the movie below.

Create a Custom AI Chatbot with Chat with RTX – YouTube

There are two types of AI models that can be used with Chat with RTX: Mistral and Llama 2.

Files in local folders can be specified as datasets. In addition to text files, it also supports PDF and DOC formats.

When I asked Chat with RTX, “What is the name of the restaurant that Sarah recommended?”, he told me the name of the restaurant by pointing to text files such as meeting records in the dataset.

You can also select YouTube URLs in your dataset. In the movie,Consumer Electronics Show 2024Nvidia inAdvertisement videoThe URL has been entered.

「Consumer Electronics Show 2024So what type of GPU has NVIDIA announced? “, replied to chat with RTX from the video, “NVIDIA announces RTX 4080 Super, RTX 4070 Ti Super, and RTX 4070 Super.''

With Chat with RTX, you can use Chat with RTX to summarize long minutes available to you, or you can extract just the items you want to find from the written content in a PDF.

Chat with RTX requires GeForce RTX 30 series or later GPU with at least 8GB of VRAM, Windows 10/11, and NVIDIA GPU driver version 551.52 or later. At the time of creating the article, it has not yet been released in Japan, and you can set a notification to start the release from below.

Build a custom LLM with Chat With RTX | NVIDIA
https://www.nvidia.com/ja-jp/ai-on-rtx/chat-with-rtx-geneative-ai/