NVIDIA introduced a series of new AI foundation models, tools, and hardware designed to bring generative and agentic AI capabilities directly to consumer PCs and enterprise systems.
At CES 2025, NVIDIA introduced a series of new AI foundation models, tools, and hardware designed to bring generative and agentic AI capabilities directly to consumer PCs and enterprise systems. These innovations, powered by NVIDIA’s latest GeForce RTX™ 50 Series GPUs and the NVIDIA NIM™ microservices, aim to redefine how developers, businesses, and enthusiasts interact with AI.
AI Foundation Models for RTX AI PCs
NVIDIA’s new foundation models, running locally on RTX AI PCs, are set to revolutionize content generation, productivity, AI development, and what Nvidia calls digital human creation. These models are powered by the GeForce RTX 50 Series GPUs, which feature up to 3,352 trillion operations per second of AI performance and 32GB of VRAM. Built on the NVIDIA Blackwell architecture, these GPUs are the first consumer hardware to support FP4 compute, doubling AI inference performance and enabling generative AI models to run locally with a smaller memory footprint.
NVIDIA NIM Microservices and AI Blueprints
NVIDIA NIM microservices allow developers and enthusiasts to easily deploy and integrate these models into workflows. They are optimized for deployment across NVIDIA GPUs, whether on PCs, workstations, or in the cloud. NIM microservices are compatible with popular AI development frameworks such as LangChain, Langflow, and the AI Toolkit for VSCode, enabling seamless integration into existing applications.
These microservices are supported by a pipeline of models from leading developers, including Black Forest Labs, Meta, Mistral, and Stability AI. Use cases span large language models (LLMs), vision-language models, image generation, speech processing, and retrieval-augmented generation (RAG).
One of the highlights is the Llama Nemotron family of open models, optimized for agentic AI tasks such as instruction following, function calling, coding, and math. For instance, the Llama Nemotron Nano model is designed to run efficiently on RTX AI PCs and workstations, enabling advanced AI capabilities like chat and coding directly on consumer hardware.
To demonstrate the potential of NIM microservices, NVIDIA previewed Project R2X, a vision-enabled PC avatar powered by NVIDIA RTX Neural Faces and Audio2Face™-3D. This avatar can assist users with tasks such as summarizing documents, managing desktop apps, and enhancing video conference calls. Project R2X connects to cloud AI services like OpenAI’s GPT4o and xAI’s Grok, as well as NIM microservices, showcasing the versatility of NVIDIA’s AI ecosystem.
NVIDIA also introduced AI Blueprints, preconfigured reference workflows that streamline agentic and generative AI application development and deployment. These Blueprints enable enterprises to build and operationalize custom AI solutions, creating data-driven AI flywheels that enhance productivity. These Blueprints leverage NIM microservices to simplify complex tasks. For example, the PDF-to-podcast blueprint extracts text, images, and tables from a PDF, generates a podcast script, and creates an audio recording using AI-generated or user-provided voice samples. Another blueprint, designed for 3D-guided generative AI, allows artists to control image generation using 3D scenes created in tools like Blender.
Blueprints demonstrate how AI can enhance creativity and productivity. They provide developers and creators with powerful tools to streamline their workflows. By running locally on RTX AI PCs, these blueprints eliminate the need for cloud-based processing, offering faster and more secure solutions.
Agentic AI and the Nemotron Model Families
NVIDIA also introduced the Llama Nemotron and Cosmos Nemotron model families, designed to advance agentic AI — a new era of AI where specialized agents collaborate to solve complex problems and automate tasks. These models are optimized for enterprise applications, including customer support, fraud detection, and supply chain management.
The Llama Nemotron models, built on the popular Llama foundation, are pruned and trained using NVIDIA’s NeMo and the latest techniques to enhance efficiency and accuracy. They are available in three sizes — Nano, Super, and Ultra — to cater to various deployment needs, from real-time applications on PCs to data-center-scale operations. These models are also customizable using NVIDIA NeMo microservices, allowing enterprises to tailor them to specific domains and use cases.
The Cosmos Nemotron models, on the other hand, focus on vision-language tasks, enabling AI agents to analyze and respond to images and videos. These models suit autonomous machines, healthcare, retail, and media applications. NVIDIA also announced Cosmos world foundation models for generating physics-aware videos, further expanding the capabilities of AI agents in robotics and autonomous vehicles.
Availability and Industry Support
NVIDIA’s NIM microservices and AI Blueprints will be available in February 2025. Initial hardware support will be provided for GeForce RTX 50 Series GPUs, select RTX 40 Series, and professional GPUs. Leading manufacturers, including Acer, ASUS, Dell, HP, Lenovo, and MSI, as well as custom system builders like Corsair and Falcon Northwest, will offer NIM-ready RTX AI PCs.
The Llama Nemotron and Cosmos Nemotron models will also be available soon as downloadable models and hosted APIs, with free access for development and research through the NVIDIA Developer Program. Enterprises can deploy these models using the NVIDIA AI Enterprise software platform, ensuring seamless integration into their workflows.
What These Announcements Mean Today
RTX AI PCs and NIM microservices bring generative and agentic AI capabilities directly to consumer PCs, making cutting-edge AI tools accessible to a broader audience. Tasks that once required powerful data centers can now be performed locally, enabling faster, more secure, and more personalized AI experiences. From creating digital humans and automating workflows to building intelligent AI agents, these tools empower developers and enthusiasts to push the boundaries of what’s possible.
In gaming, NVIDIA ACE is redefining how players interact with virtual worlds. Autonomous game characters powered by ACE bring a new level of realism and dynamism to NPCs, enabling them to perceive, plan, and act like human players. This technology is already being integrated into significant titles like PUBG: BATTLEGROUNDS and NARAKA: BLADEPOINT, where AI teammates and enemies adapt to player behavior, creating more immersive and unpredictable gameplay. Beyond gaming, ACE’s generative AI capabilities are also transforming game development, with tools like Audio2Face streamlining animation workflows and enabling lifelike character interactions. Together, these innovations signal a future where AI is seamlessly integrated into every aspect of our digital lives, from productivity to entertainment.
One of the most exciting applications of ACE is in the upcoming murder mystery game Dead Meat, where players can talk to any character using natural language. Powered by NVIDIA ACE and small language models, Dead Meat allows players to interrogate suspects, ask open-ended questions, and manipulate or charm them into revealing secrets. This level of interaction, previously only possible with human players, creates a dynamic and immersive experience in which every conversation can shape the game’s outcome.
Conclusion
NVIDIA’s announcements at CES 2025 highlight the company’s commitment to pushing the boundaries of AI innovation across industries. By introducing powerful Nemotron model families with NIM microservices AI Blueprints, NVIDIA enables developers, enterprises, and creators to unlock the full potential of generative and agentic AI. These advancements, powered by the cutting-edge GeForce RTX 50 Series GPUs, bring AI capabilities directly to consumer PCs, making them faster, more secure, and more accessible than ever before.
From revolutionizing productivity and content creation to transforming gaming with NVIDIA ACE autonomous characters, NVIDIA sets the stage for a new era of AI-driven experiences. Whether enabling lifelike NPCs in games, streamlining creative workflows, or empowering enterprises to build intelligent AI agents, NVIDIA’s innovations are reshaping how we interact with technology. As these tools and technologies become available, they promise to redefine the possibilities of AI in both personal and professional spheres, paving the way for a more intelligent and immersive future.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
