Related papers: Silicon Showdown: Performance, Efficiency, and Eco…
Analog in-memory computing (AIMC) is a promising compute paradigm to improve speed and power efficiency of neural network inference beyond the limits of conventional von Neumann-based architectures. However, AIMC introduces fundamental…
Large language models (LLMs) have been increasingly deployed as local agents on personal devices with CPUs, NPUs and integrated GPUs. However, forecasting inference performance on devices with such heterogeneity remains challenging due to…
The rapid development of large language models (LLM) has greatly enhanced everyday applications. While many FPGA-based accelerators, with flexibility for fine-grained data control, exhibit superior speed and energy efficiency compared to…
Deploying local large language models and vision-language models on edge devices requires balancing accuracy with constrained computational and energy budgets. Although graphics processors dominate modern artificial-intelligence deployment,…
Large Language Model (LLM) inference requires substantial computational resources, yet CPU-based inference remains essential for democratizing AI due to the widespread availability of CPUs compared to specialized accelerators. However,…
Large Language Models (LLMs) achieve strong performance across tasks, but face storage and compute challenges on edge devices. We propose EntroLLM, a compression framework combining mixed quantization and entropy coding to reduce storage…
Recently, large language models (LLMs) have achieved huge success in the natural language processing (NLP) field, driving a growing demand to extend their deployment from the cloud to edge devices. However, deploying LLMs on…
The rapid scaling of large language models (LLMs) has unveiled critical limitations in current hardware architectures, including constraints in memory capacity, computational efficiency, and interconnection bandwidth. DeepSeek-V3, trained…
Large language model (LLM) inference has been a prevalent demand in daily life and industries. The large tensor sizes and computing complexities in LLMs have brought challenges to memory, computing, and databus. This paper proposes a…
While Neural Processing Units (NPUs) offer high theoretical efficiency for edge AI, state-of-the-art Vision--Language Models (VLMs) tailored for GPUs often falter on these substrates. We attribute this hardware-model mismatch to two primary…
With the ubiquitous use of modern large language models (LLMs) across industries, the inference serving for these models is ever expanding. Given the high compute and memory requirements of modern LLMs, more and more top-of-the-line GPUs…
Deploying Large Language Models (LLMs) on resource-constrained edge devices faces critical bottlenecks in memory bandwidth and power consumption. While ternary quantization (e.g., BitNet b1.58) significantly reduces model size, its direct…
Scale is often attributed as one of the factors that cause an increase in the performance of LLMs, resulting in models with billion and trillion parameters. One of the limitations of such large models is the high computational requirements…
Large Language Models (LLMs) are becoming the backbone of modern cloud services, yet their inference costs are dominated by GPU energy. Unlike traditional GPU workloads, LLM inference has two stages with different characteristics: the…
This paper investigates the architectural features and performance potential of the Apple Silicon M-Series SoCs (M1, M2, M3, and M4) for HPC. We provide a detailed review of the CPU and GPU designs, the unified memory architecture, and…
Large language model (LLM) decoding is a major inference bottleneck because its low arithmetic intensity makes performance highly sensitive to memory bandwidth. 3D-stacked near-memory processing (NMP) provides substantially higher local…
Large language models (LLMs) have shown immense potential across various domains, but their high memory requirements and inference costs remain critical challenges for deployment. Post-training quantization (PTQ) has emerged as a promising…
Ternary weight quantization (e.g., BitNet b1.58) offers a promising path to mitigate the memory bandwidth bottleneck in Large Language Model (LLM) inference. However, conventional compute platforms lack native support for ternary-weight…
In-memory computing (IMC) on a monolithic chip for deep learning faces dramatic challenges on area, yield, and on-chip interconnection cost due to the ever-increasing model sizes. 2.5D integration or chiplet-based architectures interconnect…
The common assumption in on-device AI is that GPUs, with their superior parallel processing, always provide the best performance for large language model (LLM) inference. In this work, we challenge this notion by empirically demonstrating…