SA-26-112 Critical

NVIDIA Triton Inference Server: NVIDIA Triton Inference Server contains a vulnerability in the DALI backend where an attacker…

NVIDIA Triton Inference Server

Severity
Critical, CVSS 9.8
CVSS vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Record
CVE-2026-24213
Affects
NVIDIA Triton Inference Server
Known exploited
Not listed in the CISA Known Exploited Vulnerabilities catalog
Published
Record first published by the source
May 20, 2026
Source last modified
July 24, 2026

Description

NVIDIA Triton Inference Server contains a vulnerability in the DALI backend where an attacker could cause an out-of-bounds read. A successful exploit of this vulnerability might lead to code execution, data tampering, denial of service, or information disclosure.

Description quoted from the NVD record.

Who is affected

NVIDIA Triton Inference Server is software that runs AI models so that other programs and apps can send it requests and get answers back. This flaw is in a part of the product called the DALI backend, where an attacker could make the software read memory outside the area it should. No specific affected versions are named for this flaw. A successful attack might let the attacker run their own commands, change data, stop the service from working, or see information they should not. This flaw is rated Critical.

What to do

  1. If someone in your organization installed or runs NVIDIA Triton Inference Server, that person should handle this. If an IT provider or vendor runs it for you, send them this advisory and ask when the update will be done.
  2. No fixed version is named for this flaw yet, so check the vendor advisory linked on this page for a fixed release.
  3. When a fixed release is available, install it on every system that runs NVIDIA Triton Inference Server.
  4. Until then, check the vendor advisory linked on this page regularly for updates about this flaw.

Analysis written from the source record as of July 24, 2026 and reviewed before publishing.

Sources

MORE ADVISORIES

See all