NVIDIA announced new RTX technology at COMPUTEX to power AI assistants and digital humans on GeForce RTX AI laptops. The company introduced Project G-Assist, an AI assistant technology demo for PC games and apps, debuting with ARK: Survival Ascended from Studio Wildcard. Additionally, NVIDIA launched the first PC-based NIM inference microservices for the ACE digital human platform.

These advancements are supported by the NVIDIA RTX AI Toolkit, which aids developers in optimizing and deploying large generative AI models on Windows PCs. The toolkit is part of NVIDIA’s broader efforts to accelerate over 500 PC applications and games and support 200 laptop designs from various manufacturers.

Newly announced RTX AI PC laptops from ASUS and MSI feature up to GeForce RTX 4070 GPUs and power-efficient systems-on-a-chip with Windows 11 AI capabilities. These devices will receive a free update to Copilot+ experiences when available.

“NVIDIA launched the era of AI PCs in 2018 with the release of RTX Tensor Core GPUs and NVIDIA DLSS,” said Jason Paul, vice president of consumer AI at NVIDIA. “Now, with Project G-Assist and NVIDIA ACE, we’re unlocking the next generation of AI-powered experiences for over 100 million RTX AI PC users.”

Project G-Assist aims to transform gaming by offering context-aware help using generative AI. It takes voice or text inputs along with contextual information from the game screen to provide tailored responses based on a game knowledge database.

In collaboration with Studio Wildcard, Project G-Assist was demonstrated with ARK: Survival Ascended. The assistant can answer questions about creatures, items, lore, objectives, bosses, and more while personalizing responses based on the player’s session. It also optimizes gaming system performance by providing insights into metrics and adjusting settings.

NVIDIA's ACE technology is now available for RTX AI PCs through NIM inference microservices that reduce deployment times significantly. At COMPUTEX, ACE NIM debuted in a tech demo developed with Inworld AI featuring Audio2Face™ and Riva automatic speech recognition running locally on devices.

Microsoft and NVIDIA are collaborating to bring new generative AI capabilities to Windows native and web apps via GPU-accelerated small language models (SLMs). These SLMs enable content summarization, generation, task automation, and retrieval-augmented generation (RAG) capabilities that run on-device as part of Windows Copilot Runtime.

The new NVIDIA RTX AI Toolkit helps developers build application-specific models that run efficiently on PCs by customizing pretrained models using open-source QLoRa tools. Models optimized through TensorRT consume less RAM while delivering faster performance compared to their original versions.

Software partners like Adobe are integrating components of the toolkit within their creative apps to enhance performance on RTX PCs. “Adobe and NVIDIA continue to collaborate to deliver breakthrough customer experiences across all creative workflows,” said Deepa Subramaniam, vice president of product marketing at Creative Cloud at Adobe.

RTX acceleration is also being integrated into popular user interfaces such as Automatic1111’s Stable Diffusion UI. Starting this week, ComfyUI will benefit from similar enhancements resulting in significant performance improvements over existing versions.

NVIDIA’s modding platform RTX Remix facilitates remastering classic DirectX games with ray tracing features like DLSS 3.5. Since its launch earlier this year, it has seen widespread adoption among modders leading to numerous projects currently under development.

Additionally, components of the RTX Remix Toolkit will be made open source this month allowing modders greater flexibility in enhancing game assets and rendering capabilities across various formats beyond DirectX classics.

Lastly, NVIDIA announced that its popular super-resolution feature supported in major browsers will soon be available as an SDK for developers enabling integration into video editing software like DaVinci Resolve and Wondershare Filmora among others.