NVIDIA has unveiled a significant leap in AI performance, introducing foundation models that run locally on NVIDIA RTX AI PCs.
These models are powered by the new GeForce RTX 50 Series GPUs, which deliver up to 3,352 trillion operations per second, supporting generative AI applications in digital human creation, content production, and
productivity tools.
These GPUs feature 32GB of VRAM and are built on the innovative NVIDIA Blackwell architecture, with support for FP4 compute, enhancing AI inference speed by 2x and reducing the memory footprint.
AI Power for Developers
The introduction of NVIDIA NIM microservices is a game changer for AI developers.
- Microservices optimized for RTX AI PCs enable seamless deployment of innovative AI models.
- Supports a wide range of applications including language models and computer vision.
- New low-code and no-code tools like: AnythingLLM, ComfyUI, Langflow
- These tools allow both developers and enthusiasts to leverage AI models through user-friendly interfaces.
The ease of use makes it possible for anyone to create, test, and deploy AI applications, accelerating innovation across industries.
AI Blueprints to Streamline Creation
NVIDIA is also launching AI Blueprints, preconfigured workflows that enable developers to build AI-powered applications more easily. These include workflows for creating podcasts from PDFs, generating 3D-guided images, and more.
Using NIM microservices, such as Mistral-Nemo-12B-Instruct and NVIDIA Riva for text-to-speech, these blueprints simplify complex tasks and ensure that AI-powered projects are accessible to creators of all skill levels.
Starting in February, NVIDIA’s NIM-ready RTX AI PCs will be available from top manufacturers like Acer, ASUS, Dell, HP, Lenovo, and MSI, bringing this powerful technology to a wider audience.
