Not all chatbots are confined to using only one AI model. Oftentimes, chatbots are compatible with different models that offer different levels of functionality. Nvidia is experimenting with its chatbot called ChatRTX, and the company announced that it is compatible with Google’s Gemma model along with other functionalities.
Right now, much of the AI industry is riding on Nvidia’s GPUs. Top AI companies are collaborating with Nvidia and buying boatloads of its GPUs to train their AI models. This is something that pushed Nvidia to become an over $2 trillion company. This should show how high the demand is for powerful GPUs.
So, it seems fitting that the company would eventually come out with its own public chatbot to allow people to interact with the models its GPUs are training.
Nvidia’s ChatRTX is now compatible with Google Gemma
Several months ago, Google introduced Gemini, its family of powerful AI models that come in three different sizes. However, there is a massive gap between two of the models. Gemini Nano is small enough to fit on phones, but Gemini Pro, the next step up, needs to be held in massive data centers.
Google Gemma is the middleman between those two models. It’s larger and more capable than Gemini Nano, but it is still small and light enough to fit on computers. ChatRTX can now use the Gemma model. This is good news for people who are pining for what Google has to offer. ChatRTX is already compatible with models such as Meta’s Llama 2 and Mistral.
ChatRTX is getting more functionality as well
Along with getting access to Gemma, ChatRTX will soon support voice queries. This will allow you to speak directly with the chatbot using your voice. It will forgo the process of typing in your queries.
Next, ChatRTX is now compatible with the ChatGLM3 model. This is actually a bilingual model that supports both English and Chinese.
Rounding out the list, ChatRTX is now compatible with OpenAI’s Contrastive Language Image Pre-training, AKA. CLIP. This is a tool that people can use to help users train the model to recognize images.
Do you want ChatRTX?
If you’re interested in getting this chatbot, there will be some requirements. Firstly, this chatbot downloads at a whopping 36GB. As for the system, you will need Windows 11, at least 16 GB of RAM, and that 535.11 driver or later.
As for the GPU, you have a few options. You’ll either need an Nvidia GeForce RTX 30 series or 40 series GPU. If you don’t have one of those, you can use an Nvidia RTX Ampere or Ada generation GPU with at least 8GB of VRAM.