➀ Introduces new CUDA libraries for accelerated computing, offering order-of-magnitude speedups and reduced energy consumption. ➁ Includes libraries for LLM applications, data processing, and physical AI simulations. ➂ Highlights real-world examples of energy-efficient and high-speed computing in industries like recycling and video conferencing.

➀ Researchers at Central University have developed a hybrid RF-VLC technology that promises reliable indoor communication with reduced power consumption. ➁ The system integrates RF and VLC to provide both communication and illumination while consuming less power. ➂ Initial evaluations show significant energy savings and potential reduction in electromagnetic radiation.

➀ Smart manikins equipped with sensor technology and mathematical modeling help optimize office temperatures for energy efficiency and comfort. ➁ 'HVAC' and 'ANDI' manikins measure thermal radiation and humidity to create a virtual thermal model of a human workplace. ➂ The research aims to optimize building energy needs and improve conditions in operating rooms and other environments.

➀ The 'KitchenGuard' project at TU Ilmenau is developing a new sensor to make cooking more relaxed, energy-efficient, and safer using AI. ➁ The sensor enhances induction cooktops with smart cooking functions and intuitive control. ➂ The project has received a 12-month EXIST startup grant to support its innovative business idea.

1. The Fraunhofer Institute for Energy Economics and Energy System Technology (IEE) is developing models and methods to assess the impact of local restrictions on realizable wind energy capacities and yields. 2. These models help in evaluating the usability of existing and planned sites and estimating repowering potentials. 3. The project, funded by the Federal Ministry for Economic Affairs and Climate Action, aims to accelerate the expansion of wind energy by providing tools for optimal land use and early warning systems for potential shortfalls in energy targets.

1. Researchers at ETH Zurich have developed a new method to significantly reduce the amount of environmentally harmful fluorine in lithium-metal batteries, making them more sustainable. 2. This new approach involves using electrostatic attraction to control the formation of a protective layer in the battery, enhancing stability and reducing costs. 3. The method has been successfully tested in coin-sized batteries and is now being scaled up for use in pouch cells, commonly used in smartphones.

1. Nuvoton Technology introduces the Arm Cortex-M23 M2L31 microcontroller series, focusing on energy-efficient computing. 2. The series features high power efficiency and robust processing capabilities, suitable for battery management and industrial automation. 3. It includes advanced features like ReRAM and USB Type-C PD 3.0 support, enhancing sustainability and security.

1. Researchers from UC Santa Cruz have discovered a method to run large language models (LLMs) at a mere 13 watts without compromising performance. 2. The key to this efficiency is the elimination of matrix multiplication in LLM processing, which, when optimized, significantly boosts performance-per-watt. 3. The broader applicability of this approach to AI in general is yet to be determined.

1. Mobix Labs Inc. has announced a strategic partnership with TalkingHeads Wireless to develop cost-effective and energy-efficient 5G base stations. 2. THW's 5G radio solution incorporates AI technology to optimize energy consumption in towers. 3. Mobix Labs contributes with a range of products including ICs, antennas, and active optical cables and transceivers.

1. Cambridge GaN Devices (CGD) has introduced the ICeGaN™ GaN power IC, designed for high-efficiency applications in data centers, inverters, and industrial power supplies. 2. The new ICeGaN™ P2 series features a low RDS(on) of 25 mΩ, supporting multi-kW power applications. 3. CGD's Chief Commercial Officer, Andrea Bricconi, highlighted the growing demand for GaN solutions in data centers due to the increasing energy consumption from AI growth.