News from the AI & ML world
DeeperML
@developer.nvidia.com
// 40d
NVIDIA is enhancing AI capabilities on RTX AI PCs with the introduction of a new plug-in builder for G-Assist. This innovative tool allows users to customize and expand G-Assist's functionality by integrating it with various Large Language Models (LLMs) and software applications. The plug-in builder is designed to enable users to generate AI-assisted functionalities through text and voice commands, effectively transforming G-Assist from a gaming-centric AI into a versatile tool adaptable to diverse applications, both gaming-related and otherwise.
The G-Assist plug-in builder facilitates the creation of custom commands and connections to external tools through APIs, allowing different software and services to communicate with each other. Developers can leverage coding languages like JSON and Python to create and integrate tools into G-Assist. NVIDIA has provided a GitHub repository with instructions and documentation for building and customizing these plug-ins, and users can even submit their creations for potential inclusion in the repository to share new capabilities with others. Examples of plug-in capabilities include seeking advice from AI assistants like Gemini on gaming strategies and using Twitch plug-ins to monitor streamer status via voice commands.
Furthermore, NVIDIA is advancing AI research and application across industries, demonstrated by their participation in the International Conference on Learning Representations (ICLR). NVIDIA Research presented over 70 papers at ICLR, showcasing developments in areas such as autonomous vehicles, healthcare, and multimodal content creation. Notably, researchers from University College London (UCL) are leveraging NVIDIA NIM microservices to benchmark agentic capabilities of AI models in gaming environments, highlighting the role of NIM in simplifying and accelerating the evaluation of AI reasoning in complex tasks. NIM microservices enable efficient deployment and scaling of AI models, supporting various platforms and workflows, making them a versatile solution for diverse research applications.
ImgSrc: developer-blogs
References :
The G-Assist plug-in builder facilitates the creation of custom commands and connections to external tools through APIs, allowing different software and services to communicate with each other. Developers can leverage coding languages like JSON and Python to create and integrate tools into G-Assist. NVIDIA has provided a GitHub repository with instructions and documentation for building and customizing these plug-ins, and users can even submit their creations for potential inclusion in the repository to share new capabilities with others. Examples of plug-in capabilities include seeking advice from AI assistants like Gemini on gaming strategies and using Twitch plug-ins to monitor streamer status via voice commands.
Furthermore, NVIDIA is advancing AI research and application across industries, demonstrated by their participation in the International Conference on Learning Representations (ICLR). NVIDIA Research presented over 70 papers at ICLR, showcasing developments in areas such as autonomous vehicles, healthcare, and multimodal content creation. Notably, researchers from University College London (UCL) are leveraging NVIDIA NIM microservices to benchmark agentic capabilities of AI models in gaming environments, highlighting the role of NIM in simplifying and accelerating the evaluation of AI reasoning in complex tasks. NIM microservices enable efficient deployment and scaling of AI models, supporting various platforms and workflows, making them a versatile solution for diverse research applications.
ImgSrc: developer-blogs
Share:
References :
- blogs.nvidia.com: NVIDIA Research at ICLR — Pioneering the Next Wave of Multimodal Generative AI
- www.tomshardware.com: Nvidia introduces G-Assist plug-in builder, allowing its AI to integrate with LLMs and software
Classification:
- HashTags: #NVIDIA #AIPlugins #MultimodalAI
- Company: Nvidia
- Target: AI developers
- Product: G-Assist
- Feature: G-Assist
- Type: AI
- Severity: Informative