Cyber Posture

CVE-2026-24152

High

Published: 24 March 2026

Published
24 March 2026
Modified
25 March 2026
KEV Added
Patch
CVSS Score 7.8 CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
EPSS Score 0.0006 19.8th percentile
Risk Priority 16 60% EPSS · 20% KEV · 20% CVSS

Summary

CVE-2026-24152 is a high-severity Deserialization of Untrusted Data (CWE-502) vulnerability in Nvidia Megatron-Lm. Its CVSS base score is 7.8 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Malicious File (T1204.002); ranked at the 19.8th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

The strongest mitigations our analysis identified are NIST 800-53 SI-10 (Information Input Validation) and SI-2 (Flaw Remediation).

Threat & Defense at a Glance

What attackers do: exploitation maps to Malicious File (T1204.002) and 1 other technique. What defenders deploy: see the NIST 800-53 controls recommended below.
Threat & Defense Details

Mitigating Controls (NIST 800-53 r5)AI

prevent

Directly addresses the CVE by requiring identification, reporting, and timely remediation of the deserialization flaw in Megatron-LM checkpoint loading.

prevent

Mandates validation of checkpoint files as information inputs to prevent improper deserialization leading to RCE.

prevent

Requires integrity verification of checkpoint files before loading to detect and block maliciously crafted data causing code execution.

MITRE ATT&CK Enterprise TechniquesAI

T1204.002 Malicious File Execution
An adversary may rely upon a user opening a malicious file in order to gain execution.
T1059.006 Python Execution
Adversaries may abuse Python commands and scripts for execution.
Why these techniques?

Direct RCE via malicious checkpoint file (deserialization) requires user execution of crafted file in Python-based framework.

Confidence: HIGH · MITRE ATT&CK Enterprise v18.1

NVD Description

NVIDIA Megatron-LM contains a vulnerability in checkpoint loading where an Attacker may cause an RCE by convincing a user to load a maliciously crafted file. A successful exploit of this vulnerability may lead to code execution, escalation of privileges, information…

more

disclosure, and data tampering.

Deeper analysisAI

CVE-2026-24152 affects NVIDIA Megatron-LM, a framework for training large language models, specifically in its checkpoint loading functionality. The vulnerability arises from improper deserialization of untrusted data (CWE-502), enabling an attacker to achieve remote code execution (RCE) by crafting a malicious checkpoint file. Published on 2026-03-24, it carries a CVSS v3.1 base score of 7.8 (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H), indicating high severity due to substantial impacts on confidentiality, integrity, and availability.

A local attacker with low privileges can exploit this vulnerability by convincing a legitimate user to load the maliciously crafted checkpoint file into Megatron-LM. Successful exploitation grants code execution on the host system, potentially leading to privilege escalation, information disclosure, and data tampering.

Mitigation guidance is detailed in advisories including the NVIDIA security bulletin at https://nvidia.custhelp.com/app/answers/detail/a_id/5769, the NVD entry at https://nvd.nist.gov/vuln/detail/CVE-2026-24152, and the CVE record at https://www.cve.org/CVERecord?id=CVE-2026-24152. Security practitioners should consult these for patching instructions and workarounds specific to affected Megatron-LM versions.

Details

CWE(s)

Affected Products

nvidia
megatron-lm
≤ 0.15.3

CVEs Like This One

CVE-2026-24150Same product: Nvidia Megatron-Lm
CVE-2026-24151Same product: Nvidia Megatron-Lm
CVE-2025-33248Same product: Nvidia Megatron-Lm
CVE-2025-33247Same product: Nvidia Megatron-Lm
CVE-2026-24165Same vendor: Nvidia
CVE-2025-33253Same vendor: Nvidia
CVE-2025-33241Same vendor: Nvidia
CVE-2025-33252Same vendor: Nvidia
CVE-2025-33245Same vendor: Nvidia
CVE-2026-24157Same vendor: Nvidia

References