CVE-2026-94624: vLLM Denial of Service Vulnerability in P2P KV Offloading

vllm Updated 30 Sep 2026

Summary

21/09/2026
Yes i
Unrated
Not published

CVE-2026-94624: vLLM through 0.29.0 contains a denial of service vulnerability in P2P KV offloading when OffloadingConnector is configured with TieringOffloadingSpec and a peer-to-peer secondary tier. Affects vllm versions < 0.30.0. Fixed in 0.30.0. How to fix ↓

Free scan

Check if this vulnerability affects your application

Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.

Check my app free

No signup required · Results in about 60 seconds

Upload your dependency manifest or SBOM

Drag & drop your file here or

Java Javascript PHP Python Go Rust Ruby Dart C/C++ Elixir CycloneDX SPDX

Upload a supported dependency manifest or SBOM. No installation or source-code access required.

See supported files and SBOM formats
SBOM sbom.json · bom.json · bom.xml · *.spdx.json · *.cdx.json · *.cdx.xml
Java pom.xml · gradle.lockfile · buildscript-gradle.lockfile
Javascript package-lock.json · yarn.lock · pnpm-lock.yaml
PHP composer.lock
Python poetry.lock · requirements.txt · Pipfile.lock
Go go.mod
Rust Cargo.lock
Ruby Gemfile.lock
Dart pubspec.lock
C/C++ conan.lock
Elixir mix.lock
Enter your work email to receive your report link.

Verify your email to view the full report and start 30 days of continuous monitoring. No credit card required.

Sponsored Devnack Mobile apps built by Devnack Design, build and launch iOS and Android apps with a senior product team. From idea to store release Start a project

Background

vLLM is an open-source library designed for efficient machine learning model serving, particularly in scenarios requiring high throughput and low latency. It supports various offloading strategies to optimize resource usage. However, a recent vulnerability has been identified that affects its P2P KV offloading feature, potentially leading to service disruptions.

Vulnerability Detail

The vulnerability exists in vLLM versions up to 0.29.0, where the OffloadingConnector configured with TieringOffloadingSpec and a peer-to-peer secondary tier can be exploited. Attackers can manipulate the kv_transfer_params to specify arbitrary remote host and port values, resulting in the creation of unreachable peer sessions. This leads to the retention of ZeroMQ sockets until the context quota is exhausted, ultimately causing an uncaught ZMQError that crashes the EngineCore and halts all inference processes.

How To Fix

Latest published version: 0.32.0 as of 22 Oct 2026

pip install --upgrade "vllm==0.32.0"

Upgrading to it may include breaking changes; the version below is the minimum fixed release for this vulnerability.

If you need the minimum change

0.30.0

This is the first release that contains the fix.

pip install --upgrade "vllm==0.30.0"

Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →

Workaround

As a temporary workaround, avoid configuring the OffloadingConnector with arbitrary remote host and port values in kv_transfer_params. Ensure that only trusted and reachable peer addresses are used to prevent the creation of unreachable peer sessions.

Conclusion

The CVE-2026-94624 vulnerability in vLLM poses a significant risk of denial of service through improper configuration of the P2P KV offloading feature. Users are advised to update to version 0.30.0 or later to mitigate this risk and ensure the stability of their applications.

See every vulnerability recorded for vllm →

Share this vulnerability

FAQ

CVE-2026-94624 is a denial of service vulnerability in the vLLM package that can be exploited through improper configuration of the P2P KV offloading feature.

Versions of vLLM up to 0.29.0 are affected by this vulnerability. Version 0.30.0 and later contain the necessary fixes.

You can use the Vulert playground at Vulert Scanner to check if your application is affected by this vulnerability without needing to sign up.

Exploitation of this vulnerability can lead to service crashes, stopping all inference processes and potentially affecting application availability.

Yes, Vulert can monitor and alert you in real-time for open-source vulnerabilities in your software by analyzing the Pipfile.lock.

Related vulnerabilities

Recent vulnerabilities recorded for vllm.

How To Fix

Latest published version: 0.32.0 as of 22 Oct 2026

pip install --upgrade "vllm==0.32.0"

Upgrading to it may include breaking changes; the version below is the minimum fixed release for this vulnerability.

If you need the minimum change

0.30.0

This is the first release that contains the fix.

pip install --upgrade "vllm==0.30.0"

Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →

Subscribe to Vulert on Telegram
Daily digest of new CVEs, critical and high severity, affected packages and fixes — plus a weekly roundup.
Join the channel →
Help improve this record
Report an issue or send feedback