CVE-2026-24271

Deferred Deferred - Pending Action

Denial of Service in NVIDIA TensorRT-LLM

Publication date: 2026-07-14

Last updated on: 2026-07-15

Assigner: NVIDIA Corporation

Description

NVIDIA TensorRT-LLM contains a vulnerability in the OpenAI-compatible inference API, where an attacker could cause allocation of GPU resources without limits or throttling. A successful exploit of this vulnerability might lead to denial of service.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published

2026-07-14

Last Modified

2026-07-15

Generated

2026-08-04

AI Q&A

2026-07-15

EPSS Evaluated

2026-08-02

NVD

CVE-2026-24271

Affected Vendors & Products

Vendor	Product	Version / Range
nvidia	tensorrt_llm	*

Helpful Resources

Exploitability

CWE

KEV

CWE ID	Description
CWE-770	The product allocates a reusable resource or group of resources on behalf of an actor without imposing any intended restrictions on the size or number of resources that can be allocated.

Attack-Flow Graph

Executive Summary

NVIDIA TensorRT-LLM has a flaw in its OpenAI-compatible inference API that allows an attacker to trigger unlimited GPU resource allocation without any throttling or limits. This could lead to a denial of service condition by exhausting GPU resources.

Impact Analysis

If exploited, this vulnerability could cause your system to run out of GPU resources, leading to degraded performance or complete unavailability of services relying on TensorRT-LLM. This may disrupt applications such as AI inference workloads.

Compliance Impact

The vulnerability could lead to denial of service due to uncontrolled GPU resource allocation, which may impact system availability. This could affect compliance with standards requiring data availability and integrity, such as GDPR's data processing principles or HIPAA's access controls, if services become unavailable.

Mitigation Strategies

Monitor GPU resource usage for unusual spikes or unbounded allocations. Limit API request rates and enforce quotas to prevent excessive GPU consumption. Update TensorRT-LLM to the latest patched version as soon as possible.

Hi! I’m here to help you understand CVE-2026-24271. Ask me anything about the vulnerability, its impact, or mitigation strategies.

0/70