Related papers: Towards High-Goodput LLM Serving with Prefill-deco…

Large language models (LLMs) have shown remarkable performance across a wide range of applications, often outperforming human experts. However, deploying these gigantic models efficiently for diverse inference use cases requires carefully…

The recent advances in LLMs bring a strong demand for efficient system support to improve overall serving efficiency. As LLM inference scales towards multiple GPUs and even multiple compute nodes, various coordination patterns, such as…

This paper introduces LLMServingSim2.0, a system simulator designed for exploring heterogeneous hardware in large-scale LLM serving systems. LLMServingSim2.0 addresses two key limitations of its predecessor: (1) integrating hardware models…

In the rapidly evolving landscape of artificial intelligence (AI), generative large language models (LLMs) stand at the forefront, revolutionizing how we interact with our data. However, the computational intensity and memory consumption of…

High throughput serving of large language models (LLMs) requires batching sufficiently many requests at a time. However, existing systems struggle because the key-value cache (KV cache) memory for each request is huge and grows and shrinks…

Large language models (LLMs) are increasingly deployed as AI agents that operate in short reasoning-action loops, interleaving model computation with external calls. Unlike traditional chat applications, these agentic workloads require…

The proliferation of large language models (LLMs) has led to the adoption of Mixture-of-Experts (MoE) architectures that dynamically leverage specialized subnetworks for improved efficiency and performance. Despite their benefits, MoE…

General Large Language Models (LLMs) excel in reasoning, but those enhanced for translation struggle with reasoning tasks. To address this, we propose a novel translationenhanced recipe that begins with instruct models and applies…

Existing large language model (LLM) serving systems typically employ Prefill-Decode disaggregated architecture to prevent computational interference between the prefill and decode phases. However, in real-world LLM serving scenarios,…

Large Language Models (LLMs) have revolutionized numerous domains, driving the rise of Language-Model-as-a-Service (LMaaS) platforms that process millions of queries daily. These platforms must minimize latency and meet Service Level…

Speculative decoding (SD) accelerates large language model (LLM) inference by using a smaller draft model to propose multiple tokens that are verified by a larger target model in parallel. While prior work demonstrates substantial speedups…

Large language models (LLMs) have emerged due to their capability to generate high-quality content across diverse contexts. To reduce their explosively increasing demands for computing resources, a mixture of experts (MoE) has emerged. The…

Nowadays, service providers often deploy multiple types of LLM services within shared clusters. While the service colocation improves resource utilization, it introduces significant interference risks for latency-sensitive (LS)…

Deploying modern Speech Language Models (SpeechLMs) in streaming settings requires systems that provide low latency, high throughput, and strong guarantees of streamability. Existing systems fall short of supporting diverse models flexibly…

Mixture-of-Experts (MoE) models have become a dominant paradigm for scaling large language models, but their rapidly growing parameter sizes introduce a fundamental inefficiency during inference: most expert weights remain idle in GPU…

Large Language Models (LLMs) have presented impressive performance across several transformative tasks. However, it is non-trivial to efficiently utilize large-scale cluster resources to develop LLMs, often riddled with numerous challenges…

In a multi-tenant large language model (LLM) serving platform hosting diverse applications, some users may submit an excessive number of requests, causing the service to become unavailable to other users and creating unfairness. Existing…

Large Language Models (LLMs) are increasingly being integrated into various applications and services running on billions of mobile devices. However, deploying LLMs on resource-limited mobile devices faces a significant challenge due to…

Self-supervised learning (SSL) models usually require weeks of pre-training with dozens of high-end GPUs. These models typically have a multi-headed self-attention (MHSA) context encoder. However, MHSA takes quadratic time and space in the…

LLM agents have recently emerged as a powerful paradigm for solving complex tasks through planning, tool use, memory retrieval, and multi-step interaction. However, these agentic workflows often introduce substantial input-side overhead,…