Description
LLaMA-Factory is a tuning library for large language models. A remote code execution vulnerability was discovered in LLaMA-Factory versions up to and including 0.9.3 during the LLaMA-Factory training process. This vulnerability arises because the `vhead_file` is loaded without proper safeguards, allowing malicious attackers to execute arbitrary malicious code on the host system simply by passing a malicious `Checkpoint path` parameter through the `WebUI` interface. The attack is stealthy, as the victim remains unaware of the exploitation. The root cause is that the `vhead_file` argument is loaded without the secure parameter `weights_only=True`. Version 0.9.4 contains a fix for the issue.
Published: 2025-06-26
Score: 8.3 High
EPSS: 4.2% Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Analysis and contextual insights are available on OpenCVE Cloud.

Remediation

No vendor fix or workaround currently provided.

Additional remediation guidance may be available on OpenCVE Cloud.

Tracking

Sign in to view the affected projects.

Advisories
Source ID Title
EUVD EUVD EUVD-2025-19200 LLaMA-Factory allows Code Injection through improper vhead_file safeguards
Github GHSA Github GHSA GHSA-xj56-p8mm-qmxj LLaMA-Factory allows Code Injection through improper vhead_file safeguards
History

Tue, 02 Sep 2025 18:00:00 +0000

Type Values Removed Values Added
CPEs cpe:2.3:a:hiyouga:llama-factory:*:*:*:*:*:*:*:*

Thu, 26 Jun 2025 16:15:00 +0000

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'partial'}, 'version': '2.0.3'}


Thu, 26 Jun 2025 15:00:00 +0000

Type Values Removed Values Added
Description LLaMA-Factory is a tuning library for large language models. A remote code execution vulnerability was discovered in LLaMA-Factory versions up to and including 0.9.3 during the LLaMA-Factory training process. This vulnerability arises because the `vhead_file` is loaded without proper safeguards, allowing malicious attackers to execute arbitrary malicious code on the host system simply by passing a malicious `Checkpoint path` parameter through the `WebUI` interface. The attack is stealthy, as the victim remains unaware of the exploitation. The root cause is that the `vhead_file` argument is loaded without the secure parameter `weights_only=True`. Version 0.9.4 contains a fix for the issue.
Title LLaMA-Factory Remote Code Execution (RCE) Vulnerability
Weaknesses CWE-502
CWE-94
References
Metrics cvssV3_1

{'score': 8.3, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:L/A:H'}


Subscriptions

Hiyouga Llama-factory
cve-icon MITRE

Status: PUBLISHED

Assigner: GitHub_M

Published:

Updated: 2025-06-26T15:12:51.382Z

Reserved: 2025-06-24T03:50:36.795Z

Link: CVE-2025-53002

cve-icon Vulnrichment

Updated: 2025-06-26T15:08:16.516Z

cve-icon NVD

Status : Analyzed

Published: 2025-06-26T15:15:23.873

Modified: 2025-09-02T17:49:44.077

Link: CVE-2025-53002

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2025-07-06T22:16:30Z

Weaknesses