CyberRota Analysis
AI-GeneratedThe vLLM library versions from 0.12.0 to before 0.24.0 are vulnerable to a flaw that allows authorized remote users to send a specially crafted prompt in a /v1/completions request, leading to a fatal crash of the EngineCore and shutting down the server application. Organizations utilizing this library for LLM inference and serving should prioritize upgrading to version 0.24.0 to mitigate the risk of service disruption.
Public Exploit Signal
A public exploit, PoC, GitHub repository or Metasploit reference was detected for this CVE.
Note: these links are listed for security research and verification purposes only.
Original NVD Description
vLLM is a library for LLM inference and serving. From 0.12.0 to before 0.24.0, sending a pure prompt embeds payload in a /v1/completions request with a model using M-RoPE causes EngineCore to fail an assertion and fatally crash, shutting down the entire server application. Any remote user who is authorized to make a /v1/completions request can make such a request and induce a crash. This issue is fixed in version 0.24.0.
Related CVEs
Other vulnerabilities affecting the same vendor(s)