CVE-2026-24269

TensorRT-LLM server-side requests can expose information or alter data

NVIDIA · TensorRT-LLM

First documented disclosure · Review checkpoint

Summary

NVIDIA reports an SSRF weakness with potential information disclosure and data changes.

Severity and exploitation evidence

High · 7.1/10 · CVSS 3.1. Severity source.

Exploitation not established: The reviewed primary sources document a vulnerability or controlled demonstration, not confirmed malicious exploitation.

Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.

How the weakness works

The application can be induced to make unintended server-side requests; the exact path is not described publicly.

Conditions required

The CNA specifies local access, no prior privileges and no user interaction; additional prerequisites are undisclosed.

Potential impact

Potential information disclosure and data tampering, according to the vendor description.

Affected versions and configurations

1.3.0rc24, all platforms, as listed by NVIDIA.

Fix and mitigation status

1.3.0rc25.

Practical next steps

Upgrade and restrict service egress to required destinations.

Operating lesson

Network requests made on behalf of model workflows need explicit destination controls.

Evidence and limits

First verified disclosure October 27. The vendor describes data tampering but its CVSS vector assigns no integrity impact; preserve that distinction and the local attack vector.

Sources