CVE-2026-118613

Ultralytics torch_load Deserialization of Untrusted Data Remote Code Execution Vulnerability

Published Oct 28, 2026 CVSS 3.0 · 7.8 High EPSS < 1% KEV · Not listed
Description
Ultralytics torch_load Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Ultralytics. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file.

The specific flaw exists within the torch_load function. The issue results from the lack of proper validation of user-supplied data, which can result in deserialization of untrusted data. An attacker can leverage this vulnerability to execute code in the context of the current user. Was ZDI-CAN-29652.

Analysis

AI-generated

Technical impact

The flaw lies in Ultralytics’ torch_load function, which deserializes user‑supplied data without adequate validation. An attacker can supply crafted input that, when processed by torch_load, causes arbitrary code to execute in the context of the current user. This results in a full compromise of confidentiality, integrity, and availability for the affected installation because the attacker gains the privileges of that user.

Affected Systems

Ultralytics systems are affected. Specific affected product versions are not listed in the advisory, so all installations of Ultralytics are potentially vulnerable until a fix is applied.

Risk and Exploitability

The CVSS score of 7.8 classifies the issue as high severity. EPSS information is not available and the vulnerability is not listed in the CISA KEV catalog. Exploitation requires user interaction – the victim must visit a malicious page or open a malicious file – which makes remote exploitation less automated but still possible through phishing or malicious downloads. Given the high severity and the need for user action, the overall risk is moderate‑to‑high for organizations that expose Ultralytics services to external users or that allow users to open untrusted files.

Remediation

No solution or workaround provided in the CVE record.

OpenCVE Recommended Actions

  • Update Ultralytics to a version that includes the torch_load serialization fix as soon as it becomes available
  • If a patch is not yet available, avoid using torch_load on data from untrusted sources and restrict the function to trusted internal data only
  • Implement a file‑type validation or sandbox that blocks malicious files before they are passed to torch_load
  • Monitor logs for attempts to execute torch_load with unexpected input and investigate suspicious activity promptly

Generated by OpenCVE AI on October 28, 2026 at 19:00 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories available
History
Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'none', 'Technical Impact': 'total'}, 'version': '2.0.3'}

Type Values Removed Values Added
First Time appeared Ultralytics
Ultralytics ultralytics
Vendors & Products Ultralytics
Ultralytics ultralytics
Type Values Removed Values Added
Description Ultralytics torch_load Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Ultralytics. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file. The specific flaw exists within the torch_load function. The issue results from the lack of proper validation of user-supplied data, which can result in deserialization of untrusted data. An attacker can leverage this vulnerability to execute code in the context of the current user. Was ZDI-CAN-29652.
Title Ultralytics torch_load Deserialization of Untrusted Data Remote Code Execution Vulnerability
Weaknesses CWE-502
References
Metrics cvssV3_0

{'score': 7.8, 'vector': 'CVSS:3.0/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H'}

Weaknesses
  • CWE-502

    Deserialization of Untrusted Data

Sources

Status
PUBLISHED
Assigner
zdi
Published
Updated
Nov 5, 2026, 5:11 PM
Reserved
Oct 28, 2026, 4:53 PM
Status
Awaiting Analysis
Published
Oct 28, 2026, 5:18 PM
Modified
Nov 5, 2026, 5:17 PM