Related papers: A System for Microserving of LLMs

Large Language Models (LLMs) are being increasingly used across a wide range of tasks. However, their substantial computational demands raise concerns about the energy efficiency and sustainability of both training and inference. Inference,…

Given the limited performance and efficiency of on-device Large Language Models (LLMs), the collaborations between multiple LLMs enable desirable performance enhancements, in which data, tokens, and model weights could be shared across…

Advanced Large Language Models (LLMs) have achieved impressive performance across a wide range of complex and long-context natural language tasks. However, performing long-context LLM inference locally on a commodity GPU (a PC) with privacy…

Recent breakthroughs in solving reasoning, math and coding problems with Large Language Models (LLMs) have been enabled by investing substantial computation budgets at inference time. Therefore, inference speed is one of the most critical…

In recent years, large language models have demonstrated remarkable capabilities in natural language understanding and generation. However, these models often struggle with hallucinations and maintaining long term contextual relevance,…

Large language models (LLMs) have demonstrated remarkable success across various application domains, but their enormous sizes and computational demands pose significant challenges for deployment on resource-constrained edge devices. To…

Decentralized LLM inference distributes computation among heterogeneous nodes across the internet, offering a performant and cost-efficient solution, alternative to traditional centralized inference. However, the low cross-node network…

There emerges a promising trend of using large language models (LLMs) to generate code-like plans for complex inference tasks such as visual reasoning. This paradigm, known as LLM-based planning, provides flexibility in problem solving and…

LAPS identifies and disaggregates requests with different prompt lengths in LLM serving to reduce TTFT latency. While recent systems have decoupled the prefill and decode stages to improve throughput, they still rely on unified scheduling…

Modern key-value storage engines built on Log-Structured Merge-trees (LSM-trees), such as RocksDB and LevelDB, rely heavily on the performance of their compaction operations, which are impacted by a complex set of interdependent…

Modern LLM serving is no longer homogeneous or monolithic. Production systems now combine disaggregated execution, complex parallelism, runtime optimizations, and stateful workloads such as reasoning, agents, and RL rollouts. Simulation is…

Serverless computing has emerged as a compelling solution for cloud-based model inference. However, as modern large language models (LLMs) continue to grow in size, existing serverless platforms often face substantial model startup…

Large language model (LLM) serving is becoming an increasingly important workload for cloud providers. Based on performance SLO requirements, LLM inference requests can be divided into (a) interactive requests that have tight SLOs in the…

Large language model (LLM) inference is limited by high computational cost and memory bandwidth demands, making deployment on heterogeneous many-core processors challenging. Taking the MT-3000 processor used in the Tianhe supercomputer as…

KV cache has traditionally been stored in GPU memory to accelerate the decoding phase of large language model (LLM) inference. However, it is increasingly necessary to move KV caches outside GPU devices, to enable cache reuse across…

Multimodal large language models (MLLMs) extend LLMs with visual understanding through a three-stage pipeline: multimodal preprocessing, vision encoding, and LLM inference. While these stages enhance capability, they introduce significant…

Large Language Models (LLMs) process every token through all layers of a transformer stack, causing wasted computation on simple queries and insufficient flexibility for harder ones that need deeper reasoning. Adaptive-depth methods can…

Efficient scheduling of LLM inference tasks is essential for achieving low latency and high throughput, particularly with the growing use of reasoning-capable LLMs. Traditional strategies like First-Come-First-Serve (FCFS) often suffer from…

Efficient inference for on-device Large Language Models (LLMs) remains challenging due to limited hardware resources and the high cost of the prefill stage, which processes the full input context to construct Key-Value (KV) caches. We…

We study serving retrieval models, specifically late interaction models like ColBERT, to many concurrent users at once and under a small budget, in which the index may not fit in memory. We present ColBERT-serve, a novel serving system that…