Wed Jan 08 13:45:14 UTC 2025: ## NVIDIA Unveils Local Generative AI on New RTX 50 Series GPUs at CES

**Las Vegas, NV – January [Date of Release]** – NVIDIA today announced a significant leap in bringing the power of generative AI to consumers with its new NVIDIA NIM™ microservices, running locally on NVIDIA RTX™ AI PCs. This breakthrough is powered by the newly released GeForce RTX™ 50 Series GPUs, boasting up to 3,352 trillion operations per second of AI performance and 32GB of VRAM.

The RTX 50 Series, built on the NVIDIA Blackwell architecture, is the first consumer GPU to support FP4 compute, doubling AI inference performance and enabling generative AI models to run locally with a smaller memory footprint. This makes advanced AI capabilities accessible to a wider audience, empowering both developers and enthusiasts.

NVIDIA’s NIM microservices, offered in partnership with leading model developers like Black Forest Labs, Meta, Mistral, and Stability AI, provide easy access to a range of foundation models for tasks including large language models (LLMs), image generation, and more. These microservices are designed to work seamlessly with low-code/no-code tools, simplifying the development process.

To further streamline AI development, NVIDIA introduced AI Blueprints, pre-configured workflows for tasks like creating podcasts from PDFs and generating images guided by 3D scenes. These blueprints utilize NIM microservices for various functions, including text-to-speech and PDF extraction.

Demonstrating the potential of this technology, NVIDIA previewed Project R2X, a vision-enabled PC avatar capable of assisting with tasks like summarizing documents and managing video calls. The avatar utilizes NVIDIA RTX Neural Faces and Audio2Face™-3D for realistic rendering and animation.

“AI is advancing at light speed,” said Jensen Huang, founder and CEO of NVIDIA. “NIM microservices and AI Blueprints give PC developers and enthusiasts the building blocks to explore the magic of AI.”

Major PC manufacturers, including Acer, ASUS, Dell, GIGABYTE, HP, Lenovo, MSI, Razer, and Samsung, are launching NIM-ready RTX AI PCs featuring the GeForce RTX 50 Series GPUs. The NIM microservices will be available starting in February, initially supporting GeForce RTX 50 Series, RTX 4090, 4080, and NVIDIA RTX 6000 and 5000 professional GPUs. Additional GPU support will follow.

Read More