vLLM vulnerabilities

Product

vLLM

Vendor

vLLM project

Kind

Inference server

Securaity has published 67 advisories for vLLM: 1 Critical, 25 High, 40 Medium, 1 Low.

vLLM is an inference server, which is software that loads AI models and answers requests to them from other programs or users. It is made and maintained by the vLLM project, an open-source project. IT staff or developers typically install and run vLLM when their organization hosts AI models itself instead of using a vendor's cloud service. It runs on a server or computer that the organization itself controls.

RSS feed for vLLM