Memory Allocation Vulnerability in NVIDIA Triton Inference Server for Windows and Linux
CVE-2025-23331

7.5HIGH

Key Information:

Vendor

Nvidia
Status
Triton Inference Server
Vendor
CVE Published:
6 August 2025

What is CVE-2025-23331?

The NVIDIA Triton Inference Server, used on both Windows and Linux platforms, has a vulnerability that allows an attacker to send an invalid request with an excessively large memory allocation size. This can result in a segmentation fault, potentially leading to a denial of service. Users of Triton Inference Server should take precautions to mitigate this issue, as an exploit could disrupt service availability.

Affected Version(s)

Triton Inference Server Windows All versions prior to 25.06

References

https://nvd.nist.gov/vuln/detail/CVE-2025-23331
https://www.cve.org/CVERecord?id=CVE-2025-23331
https://nvidia.custhelp.com/app/answers/detail/a_id/5687

CVSS V3.1

Score:
7.5
Severity:
HIGH
Confidentiality:
None
Integrity:
None
Availability:
None
Attack Vector:
Network
Attack Complexity:
Low
Privileges Required:
None
User Interaction:
None
Scope:
Unchanged

Timeline

  • Vulnerability published

  • Vulnerability Reserved

JSON
.
CVE-2025-23331 : Memory Allocation Vulnerability in NVIDIA Triton Inference Server for Windows and Linux