[2606.07362] Breaking the Ice: Analyzing Cold Start Latency in vLLM

-->

Computer Science > Machine Learning

arXiv:2606.07362 (cs)

[Submitted on 5 Jun 2026]

Title:Breaking the Ice: Analyzing Cold Start Latency in vLLM

Authors:Huzaifa Shaaban Kabakibo, Animesh Trivedi, Lin Wang<br>View a PDF of the paper titled Breaking the Ice: Analyzing Cold Start Latency in vLLM, by Huzaifa Shaaban Kabakibo and 2 other authors

View PDF<br>HTML (experimental)

Abstract:As scalable inference services become popular, the cold start latency of an inference engine becomes important. Today, vLLM has evolved into the de facto inference engine of choice for many inference workloads. Although popular, due to its complexity and rapid evolution, there has not been a systematic study of its startup latency. With major architectural innovations such as the V1 API and the introduction of this http URL, this paper presents the first detailed performance characterization of vLLM startup latency. We break down the startup process into six foundational steps and demonstrate that it is predominantly CPU bound. Each step exhibits consistent and interpretable scaling trends with respect to model-level and system-level parameters, enabling fine-grained attribution of latency sources. Building on these insights, we develop a lightweight analytical model that accurately predicts vLLM startup latency for a given hardware configuration, providing actionable guidance for resource planning in large-scale inference environments. All benchmarking datasets, analysis tools, and prediction scripts are open sourced at this https URL.

Subjects:

Machine Learning (cs.LG)

Cite as:<br>arXiv:2606.07362 [cs.LG]

(or<br>arXiv:2606.07362v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2606.07362

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

Journal reference:<br>Proceedings of the 9th MLSys Conference, Bellevue, WA, USA, 2026

Submission history<br>From: Huzaifa Shaaban Kabakibo [view email]<br>[v1]<br>Fri, 5 Jun 2026 15:07:22 UTC (250 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled Breaking the Ice: Analyzing Cold Start Latency in vLLM, by Huzaifa Shaaban Kabakibo and 2 other authors<br>View PDF<br>HTML (experimental)<br>TeX Source

view license

Current browse context:

cs.LG

next >

new<br>recent<br>| 2026-06

Change to browse by:

cs

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

IArxiv recommender toggle

IArxiv Recommender<br>(What is IArxiv?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)