VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kamahori, Keisuke, Li, Shihang, Peter, Simon, Kasikci, Baris
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918488070684672
author Kamahori, Keisuke
Li, Shihang
Peter, Simon
Kasikci, Baris
author_facet Kamahori, Keisuke
Li, Shihang
Peter, Simon
Kasikci, Baris
contents For years, we have built LLM serving systems like any other critical infrastructure: a single general-purpose stack, hand-tuned over many engineer-years, meant to support every model and workload. In this paper, we take the opposite bet: a multi-agent loop that automatically synthesizes bespoke serving systems for different usage scenarios. We propose VibeServe, the first agentic loop that generates entire LLM serving stacks end-to-end. VibeServe uses an outer loop to plan and track the search over system designs, and an inner loop to implement candidates, check correctness, and measure performance on the target benchmark. In the standard deployment setting, where existing stacks are highly optimized, VibeServe remains competitive with vLLM, showing that generation-time specialization need not come at the cost of performance. More interestingly, in non-standard scenarios, VibeServe outperforms existing systems by exploiting opportunities that generic systems miss in six scenarios involving non-standard model architectures, workload knowledge, and hardware-specific optimizations. Together, these results suggest a different point in the design space for infrastructure software: generation-time specialization rather than runtime generality. Code is available at https://github.com/uw-syfi/vibe-serve.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06068
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?
Kamahori, Keisuke
Li, Shihang
Peter, Simon
Kasikci, Baris
Artificial Intelligence
Distributed, Parallel, and Cluster Computing
For years, we have built LLM serving systems like any other critical infrastructure: a single general-purpose stack, hand-tuned over many engineer-years, meant to support every model and workload. In this paper, we take the opposite bet: a multi-agent loop that automatically synthesizes bespoke serving systems for different usage scenarios. We propose VibeServe, the first agentic loop that generates entire LLM serving stacks end-to-end. VibeServe uses an outer loop to plan and track the search over system designs, and an inner loop to implement candidates, check correctness, and measure performance on the target benchmark. In the standard deployment setting, where existing stacks are highly optimized, VibeServe remains competitive with vLLM, showing that generation-time specialization need not come at the cost of performance. More interestingly, in non-standard scenarios, VibeServe outperforms existing systems by exploiting opportunities that generic systems miss in six scenarios involving non-standard model architectures, workload knowledge, and hardware-specific optimizations. Together, these results suggest a different point in the design space for infrastructure software: generation-time specialization rather than runtime generality. Code is available at https://github.com/uw-syfi/vibe-serve.
title VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?
topic Artificial Intelligence
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2605.06068