CVE-2025-48956
Unknown Unknown - Not Provided
BaseFortify

Publication date: 2025-08-21

Last updated on: 2025-10-09

Assigner: GitHub, Inc.

Description
vLLM is an inference and serving engine for large language models (LLMs). From 0.1.0 to before 0.10.1.1, a Denial of Service (DoS) vulnerability can be triggered by sending a single HTTP GET request with an extremely large header to an HTTP endpoint. This results in server memory exhaustion, potentially leading to a crash or unresponsiveness. The attack does not require authentication, making it exploitable by any remote user. This vulnerability is fixed in 0.10.1.1.
CVSS Scores
EPSS Scores
Probability:
Percentile:
Meta Information
Published
2025-08-21
Last Modified
2025-10-09
Generated
2026-05-27
AI Q&A
2025-08-21
EPSS Evaluated
2026-05-25
NVD
EUVD
Affected Vendors & Products
Showing 1 associated CPE
Vendor Product Version / Range
vllm vllm From 0.1.0 (inc) to 0.10.1.1 (exc)
Helpful Resources
Exploitability
CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-400 The product does not properly control the allocation and maintenance of a limited resource.
Attack-Flow Graph
AI Powered Q&A
Can you explain this vulnerability to me?

This vulnerability in vLLM allows an attacker to cause a Denial of Service (DoS) by sending a single HTTP GET request with an extremely large header to the server. This large header causes the server's memory to be exhausted, which can lead to the server crashing or becoming unresponsive. The attack can be performed remotely without any authentication.


How can this vulnerability impact me? :

The vulnerability can impact you by causing the vLLM server to crash or become unresponsive due to memory exhaustion triggered by a malicious HTTP request. This results in denial of service, potentially disrupting availability of services relying on vLLM.


What immediate steps should I take to mitigate this vulnerability?

The immediate step to mitigate this vulnerability is to upgrade vLLM to version 0.10.1.1 or later, where the issue is fixed. Additionally, consider implementing network-level protections such as limiting the size of HTTP headers accepted by the server to prevent memory exhaustion from extremely large headers.


Ask Our AI Assistant
Need more information? Ask your question to get an AI reply (Powered by our expertise)
0/70
EPSS Chart