NVIDIA Triton Inference Server Out-of-Bounds Write Vulnerability in Python Backend Allowing Remote Code Execution

A vulnerability exists in NVIDIA Triton Inference Server for Windows and Linux, specifically within the Python backend. This vulnerability allows an attacker to cause an out-of-bounds write by sending a crafted request. Exploitation of this issue could lead to remote code execution, denial of service, data tampering, or information disclosure.

Impact

Successful exploitation could result in remote code execution, denial of service, data tampering, or unauthorized information disclosure.

Remediation

Users are advised to update to version 25.07 or later. The latest release can be downloaded from the Triton Inference Server Releases page on GitHub.