Security Advisories
CVEs affecting projects tracked on Release Alert, from NVD & OSV.
CVE-2025-48956
7.5
HIGHCVSS v3
Published
August 21, 2025
Affected
3 projects
Assigned by
GitHub
Severity scale
010
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.
vllm-project/vllm
GitHubA high-throughput and memory-efficient inference and serving engine for LLMs
GitHubA high-throughput and memory-efficient inference and serving engine for LLMs>= 0.1.0, < 0.10.1.1NVD
vllmPython Package IndexA high-throughput and memory-efficient inference and serving engine for LLMs
Python Package IndexA high-throughput and memory-efficient inference and serving engine for LLMs>= 0.1.0, < 0.10.1.1NVD
vllm
NPM
