Serving Large Language Models (LLMs) is critical for AI-powered applications, yet it demands substantial computational resources, particularly in memory bandwidth and computational throughput. Low-precision computation has emerged as a key technique to improve efficiency while reducing resource consumption. Existing approaches for generating low-precision kernels are limited to weight bit widths that are powers of two and suffer from suboptimal performance because of high-level GPU programming abstractions. These abstractions restrict critical optimizations, such as fine-grained register management and optimized memory access patterns, that are essential for efficient low-precision computations. In this paper, we introduce Tilus, a domain-specific language designed for General-Purpose GPU (GPGPU) computing that supports low-precision data types with arbitrary bit widths from 1 to 8 while maintaining GPU programmability. Tilus features a thread-block-level programming model, a hierarchical memory space, a novel algebraic layout system, and extensive support for diverse low-precision data types. Tilus programs are compiled into highly efficient GPU programs through automatic vectorization and instruction selection. Extensive experiments demonstrate that Tilus efficiently supports a full spectrum of low-precision data types, and outperforms state-of-the-art low-precision kernels. Compared to existing compilers such as Triton and Ladder, as well as hand-optimized kernels such as QuantLLM and Marlin, Tilus achieves performance improvements of: 1.75x, 2.61x, 1.29x and 1.03x, respectively. We open-source Tilus at https://github.com/NVIDIA/tilus.

The COVID-19 pandemic caused tremendous economic and social disruption devastating people’s livelihoods and wellbeing. The World Health Organisation (WHO) recommended people engage in enjoyable activities such as videogames, helping them relax during difficult life situations. Using Self-Determination Theory as a lens, this interview study (N = 11) examined how people’s videogame behaviours were influenced by life context, and their impact on wellbeing during COVID-19. Findings established that videogames were a compensatory response to the stressors of the pandemic. Playing videogames satisfied participants’ needs for social connection, serving as a coping mechanism to distract them for a short time; following an initial prioritisation of gaming and spike in gaming hours, many returned to their regular gaming behaviours. This study contributes to the games for health literature, identifying the effects of videogames on wellbeing, and provides much needed knowledge for health practitioners and families on how videogames are used and their impact on players’ wellbeing during the pandemic as a societal stressful life event. The findings have clinical and policy implications both to identify a positive tool, and to dispel negative stigmas surrounding videogames.