CyberRota Analysis
AI-GeneratedNVIDIA TensorRT-LLM is vulnerable to a heap-based buffer overflow during tensor deserialization, which could allow an attacker to exploit the flaw for information disclosure, data tampering, or denial of service. Organizations utilizing TensorRT-LLM should prioritize patching this vulnerability to mitigate potential risks to their systems and data integrity.
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
NVIDIA TensorRT-LLM for any platform contains a vulnerability in tensor deserialization, where an attacker could cause a heap based buffer overflow. A successful exploit of this vulnerability might lead to information disclosure, data tampering, or denial of service.