Critical Remote Code Execution Vulnerability in vLLM Due to JPEG2000 Decoder Flaw
Summary
GHSA-4r2x-xpjr-7cvv: vLLM has RCE In Video Processing. Affects vllm versions >= 0.8.3, < 0.14.1. Fixed in 0.14.1. How to fix ↓
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.
No signup required · Results in about 60 seconds
Background
vLLM is an open-source library designed for efficient video processing and model serving, particularly in AI applications. It allows users to interact with video models through an API, enabling functionalities like video analysis and generation. However, a critical vulnerability has been identified in versions of vLLM prior to 0.14.1, which exposes it to remote code execution (RCE) attacks when handling video URLs.
Vulnerability Detail
The vulnerability arises from a chain of issues involving an info leak and a heap overflow in the JPEG2000 decoder used by OpenCV and FFmpeg. Specifically, the decoder mishandles color channel remapping, allowing an attacker to send a specially crafted video URL that triggers a heap overflow. This overflow can overwrite critical memory structures, leading to arbitrary code execution on the server. The vulnerability is particularly severe as it requires no authentication, making it exploitable even in default configurations.
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
0.14.1
This is the first release that contains the fix.
pip install --upgrade "vllm==0.14.1"Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →
Workaround
As a temporary workaround, avoid using vLLM to process video URLs until you can update to the patched version. Additionally, consider implementing authentication mechanisms to restrict access to the API, thereby reducing the risk of exploitation.
Conclusion
In summary, a critical RCE vulnerability exists in vLLM due to flaws in the JPEG2000 decoder used for video processing. Users are strongly advised to update to the latest version to mitigate risks. For those unable to update immediately, implementing access controls and avoiding video processing features can help reduce exposure.
See every vulnerability recorded for vllm →
FAQ
The CVSS score of this vulnerability is 9.8, indicating a critical level of severity.
You can use Vulert's playground at Vulert Scanner to check if your application is affected by this vulnerability without needing to sign up.
Versions of vLLM from 0.8.3 to below 0.14.1 are affected by this vulnerability.
If you cannot update immediately, avoid using the video processing features of vLLM and implement access controls to limit exposure.
Exploitation of this vulnerability could allow an attacker to execute arbitrary commands on the server, potentially leading to data breaches or service disruptions.
References
Related vulnerabilities
Recent vulnerabilities recorded for vllm.
