NVIDIA’s upcoming Global Technology Conference (GTC) 2025 is set to demonstrate revolutionary advances in AI computing through RTX AI PCs and workstations, with a focus on local AI deployment and development.

The event, scheduled for March 17-21 at the San Jose Convention Center, will bring together leading experts to share insights on the latest AI technologies and implementations.

The conference arrives at a pivotal moment as generative AI continues to reshape the computing landscape, particularly in how AI models are built, trained, and optimized on personal computers and workstations.

With over 100 million GeForce RTX and NVIDIA RTX GPU users worldwide, the platform represents a significant opportunity for developers to deploy new AI applications and features.

A key highlight of the conference will be the presentation on building digital humans, chatbots, and AI-generated podcasts. Annamalai Chockalingam, NVIDIA’s senior product manager, will demonstrate comprehensive tools designed to streamline development and accelerate AI-enabled applications deployment.

The event will address crucial developments in language model implementation, particularly the balance between large language models (LLMs) and small language models for specific applications. Oluwatobi Olabiyi, senior engineering manager at NVIDIA, will present techniques for creating and training small language models optimized for on-device performance.

Windows workstation optimization will receive special attention, with sessions dedicated to exploring best practices for AI workload enhancement. NVIDIA software engineers will present detailed strategies for model quantization, inference pipeline improvements, and hardware-aware tuning across various computing components.

Dell Pro Max PCs and Z by HP’s AI solutions will demonstrate how their integration with NVIDIA RTX GPUs enables secure, on-premises AI development while maintaining data control and IP protection.

In conjunction with the conference, NVIDIA is rolling out the initial public beta of NIM microservices, featuring the Llama 3.1 LLM, NVIDIA Riva Parakeet for automatic speech recognition, and YOLOX for computer vision. These optimized, prepackaged models for generative AI are designed to facilitate easier PC development through standard APIs.

GTC 2025 will feature over 1,000 sessions, 300+ exhibits, and technical training opportunities. The conference represents a comprehensive showcase of AI advancement and its practical applications across industries.

The event underscores NVIDIA’s commitment to democratizing AI development through powerful local computing solutions, marking a significant step forward in making advanced AI capabilities more accessible to developers and enterprises worldwide.

News Source: https://blogs.nvidia.com/blog/rtx-ai-garage-gtc-2025-sessions/