Expanding its AI ecosystem, NVIDIA has introduced "Chat with RTX", a chatbot for Windows PCs that is powered by TensorRT-LLM & available for free on the latest RTX GPUs.
The utility of the "Chat with RTX" chatbot is very simple, it is designed as a localized system which means that you will have a personalized GPT chatbot available to you all the time on your PC without the need to go online. Chat with RTX can be fully personalized by utilizing a dataset that is available locally on your PC and the best part is that it runs across almost all RTX 40 & RTX 30 GPUs.
Starting with the details, Chat with RTX leverages NVIDIA's TensorRT-LLM & Retrieval Augmented Generated (RAG) software which was announced for Windows PCs last year & takes full advantage of the RTX acceleration available on RTX hardware to deliver the best possible experience to users. Once again, the application is supported across all GeForce RTX 30 & 40 GPUs with at least 8 GB of video memory.
After downloading "Chat with RTX" for free, users can connect it to a local dataset available on the PC (.txt, .pdf, .doc, .docx, .xml) and connect it to a large language model such as Mistral and Llama 2. You can also add specific URLs for example for YouTube videos or entire playlists to further enhance the dataset search results. After connecting, users can then use Chat With RTX the same way as they would use ChatGPT by running different queries but the results generated will be based entirely on the specific dataset, giving you better responses compared to online methods.https://cdn.wccftech.com/wp-content/uploads/2024/02/chat-with-rtx-demo-looping-video.mp4
Having an NVIDIA RTX GPU that supports TensorRT-LLM means that you will have all your data and projects available locally ratherRead more on wccftech.com