ghsa-53q9-r3pm-6pq6
Vulnerability from github
Published
2025-04-18 15:19
Modified
2025-05-28 19:58
Severity ?
9.3 (Critical) - CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
Summary
PyTorch: `torch.load` with `weights_only=True` leads to remote code execution
Details

Description

I found a Remote Command Execution (RCE) vulnerability in PyTorch. When loading model using torch.load with weights_only=True, it can still achieve RCE.

Background knowledge

https://github.com/pytorch/pytorch/security As you can see, the PyTorch official documentation considers using torch.load() with weights_only=True to be safe. image Since everyone knows that weights_only=False is unsafe, so they will use the weights_only=True to mitigate the seucirty issue. But now, I just proved that even if you use weights_only=True, it can still achieve RCE.

Credit

This vulnerability was found by Ji'an Zhou.

Show details on source website

To clipboard 

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 2.5.1"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "torch"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "2.6.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "pytorch"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "last_affected": "2.5.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2025-32434"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-502"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2025-04-18T15:19:28Z",
    "nvd_published_at": "2025-04-18T16:15:23Z",
    "severity": "CRITICAL"
  },
  "details": "# Description\nI found a Remote Command Execution (RCE) vulnerability in PyTorch. When loading model using torch.load with weights_only=True, it can still achieve RCE.  \n\n# Background knowledge\nhttps://github.com/pytorch/pytorch/security \n As you can see, the PyTorch official documentation considers using `torch.load()` with `weights_only=True` to be safe.\n![image](https://github.com/user-attachments/assets/fdaa8520-d66a-473a-ab1f-163d793de298)\nSince everyone knows that weights_only=False is unsafe, so they will use the  weights_only=True to mitigate the seucirty issue.\nBut now, I just proved that even if you use weights_only=True, it can still achieve RCE.\n\n# Credit\nThis vulnerability was found by Ji\u0027an Zhou.",
  "id": "GHSA-53q9-r3pm-6pq6",
  "modified": "2025-05-28T19:58:26Z",
  "published": "2025-04-18T15:19:28Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/pytorch/pytorch/security/advisories/GHSA-53q9-r3pm-6pq6"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-32434"
    },
    {
      "type": "WEB",
      "url": "https://github.com/pytorch/pytorch/commit/8d4b8a920a2172523deb95bf20e8e52d50649c04"
    },
    {
      "type": "WEB",
      "url": "https://github.com/pypa/advisory-database/tree/main/vulns/torch/PYSEC-2025-41.yaml"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/pytorch/pytorch"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "PyTorch: `torch.load` with `weights_only=True` leads to remote code execution"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
  • Confirmed: The vulnerability is confirmed from an analyst perspective.
  • Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
  • Patched: This vulnerability was successfully patched by the user reporting the sighting.
  • Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
  • Not confirmed: The user expresses doubt about the veracity of the vulnerability.
  • Not patched: This vulnerability was not successfully patched by the user reporting the sighting.