CVE-2026-12484: Keras TorchModuleWrapper Can Deserialize Unsafe PyTorch Vulnerability

HIGH
Advisory Identifier: GHSA-v2w2-w228-c444
Share
CVSS 7.8
CWE-502
Published
Advisory Summary

CVE-2026-12484 is a HIGH (CVSS 7.8) torchmodulewrapper can deserialize unsafe pytorch vulnerability (CWE-502: Deserialization of Untrusted Data) affecting Keras libraries in the PyPI ecosystem. The issue involves torchmodulewrapper can deserialize unsafe pytorch pickle data. Remediation is available by upgrading affected packages to version 3.12.3, 3.15.0.

CVE-2026-12484 at a Glance

CVE Identifier CVE-2026-12484
GHSA Identifier GHSA-v2w2-w228-c444
Severity HIGH
CVSS v3.1 Base Score 7.8
CWE Classification CWE-502: Deserialization of Untrusted Data
Affected Product Keras
Package Ecosystem PyPI
Published Date
Last Updated

Security Impact

CVSS: 7.8

The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.0/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H).

Attack Vector Local (AV:L)
Attack Complexity Low (AC:L)
Privileges Required None (PR:N)
User Interaction Required (UI:R)
Scope Unchanged (S:U)
Confidentiality High (C:H)
Integrity High (I:H)
Availability High (A:H)

How to Fix CVE-2026-12484

To remediate this vulnerability, upgrade all affected Keras dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.

keras fixed: 3.12.3 
pip install keras>=3.12.3
keras fixed: 3.15.0 
pip install keras>=3.15.0
Verify Direct and Transitive Dependencies

Run the following command in your project directory to verify that all resolved dependencies use the patched release: 

pip show keras

How to Check if You Are Affected

To determine whether your Python environment is affected by CVE-2026-12484, inspect your installed packages using pip:

pip show keras

Analyze the terminal output to locate any version of Keras in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 3.12.3), your application is affected.

Note: The pip show keras command checks currently installed packages in your active Python environment.

Technical Details

Vulnerability Mechanism & Technical Analysis

This vulnerability is classified under CWE-502: Deserialization of Untrusted Data, where the product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid. In Keras, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.

Source Advisory Statement

A vulnerability in keras-team/keras version 3.15.0 allows unsafe deserialization of attacker-controlled PyTorch pickle data through the public `keras.layers.TorchModuleWrapper.from_config` method. This method invokes `torch.load(..., weights_only=False)` without requiring an explicit unsafe opt-in, such as a `safe_mode=False` parameter. When called outside a `SafeModeScope(True)` context, the absence of an ambient safe mode state permits unsafe deserialization by default. This issue can lead to arbitrary code execution if untrusted Keras layer configurations are processed using this method. The vulnerability arises because the method does not enforce safe deserialization practices unless explicitly guarded by Keras safe mode.

CVE-2026-12484 Affected Keras Versions

2 component(s)
Fixed Versions:
3.12.33.15.0
Product / Package Ecosystem Affected Versions Fixed Version
keras
pypi/keras
PyPI
< 3.12.3
3.12.3
keras
pypi/keras
PyPI
≥ 3.13.0
< 3.15.0
3.15.0

References

Keras Source Repository REPOSITORY
https://github.com/keras-team/keras/pull/23165
NVD — CVE-2026-12484 NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-12484
Keras Source Repository REPOSITORY
https://github.com/keras-team/keras/pull/23048
Security Fix Commit FIX COMMIT
https://github.com/keras-team/keras/commit/55888d3becbdbb45dc16a55b489900f911e2dde5
Security Fix Commit FIX COMMIT
https://github.com/keras-team/keras/commit/d338a45204bdc787c8b3c4a9b82c1911cd52dedf
Keras Source Repository REPOSITORY
https://github.com/keras-team/keras
Keras Release RELEASE
https://github.com/keras-team/keras/releases/tag/v3.12.3
Keras Release RELEASE
https://github.com/keras-team/keras/releases/tag/v3.15.0
huntr.com/bounties/ab14df49-1... WEB
https://huntr.com/bounties/ab14df49-13b5-4442-b754-3189430bfa28

Frequently Asked Questions

What is CVE-2026-12484?
CVE-2026-12484 is a security vulnerability in Keras involving TorchModuleWrapper can deserialize unsafe PyTorch pickle data under CWE-502: Deserialization of Untrusted Data.
Which Keras versions are affected?
Affected versions include releases prior to 3.12.3, 3.15.0 across affected packages: keras.
What version fixes CVE-2026-12484?
Upgrading to 3.12.3, 3.15.0 (or the latest patched release for your respective branch) resolves the vulnerability.
Is keras affected?
Yes, keras is affected in releases prior to 3.12.3, 3.15.0. Upgrade your dependency to 3.12.3, 3.15.0 or later.
What is the CVSS score of CVE-2026-12484?
CVE-2026-12484 has a CVSS v3.1 base score of 7.8 (HIGH) with vector string CVSS:3.0/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H.
What CWE is associated with CVE-2026-12484?
CVE-2026-12484 is classified under CWE-502: Deserialization of Untrusted Data (The product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.).
How can I check PyPI dependencies for CVE-2026-12484?
Run the dependency tree command: 'pip show keras'. This command lists direct and transitive dependencies in your build configuration so you can verify if a vulnerable version is resolved.
Machine-readable OSV JSON OSV v1.5 Schema
Normalized schema output
View Raw JSON 
{
  "schema_version" : "1.4.0",
  "id" : "GHSA-v2w2-w228-c444",
  "modified" : "2026-08-07T20:33:49Z",
  "published" : "2026-07-19T21:30:20Z",
  "aliases" : [ "CVE-2026-12484" ],
  "summary" : "Keras: TorchModuleWrapper can deserialize unsafe PyTorch pickle data",
  "details" : "A vulnerability in keras-team/keras version 3.15.0 allows unsafe deserialization of attacker-controlled PyTorch pickle data through the public `keras.layers.TorchModuleWrapper.from_config` method. This method invokes `torch.load(..., weights_only=False)` without requiring an explicit unsafe opt-in, such as a `safe_mode=False` parameter. When called outside a `SafeModeScope(True)` context, the absence of an ambient safe mode state permits unsafe deserialization by default. This issue can lead to arbitrary code execution if untrusted Keras layer configurations are processed using this method. The vulnerability arises because the method does not enforce safe deserialization practices unless explicitly guarded by Keras safe mode.",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.0/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H"
  } ],
  "affected" : [ {
    "package" : {
      "ecosystem" : "PyPI",
      "name" : "keras"
    },
    "ranges" : [ {
      "type" : "ECOSYSTEM",
      "events" : [ {
        "introduced" : "0"
      }, {
        "fixed" : "3.12.3"
      } ]
    } ]
  }, {
    "package" : {
      "ecosystem" : "PyPI",
      "name" : "keras"
    },
    "ranges" : [ {
      "type" : "ECOSYSTEM",
      "events" : [ {
        "introduced" : "3.13.0"
      }, {
        "fixed" : "3.15.0"
      } ]
    } ]
  } ],
  "references" : [ {
    "type" : "ADVISORY",
    "url" : "https://nvd.nist.gov/vuln/detail/CVE-2026-12484"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/pull/23048"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/pull/23165"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/commit/55888d3becbdbb45dc16a55b489900f911e2dde5"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/commit/d338a45204bdc787c8b3c4a9b82c1911cd52dedf"
  }, {
    "type" : "PACKAGE",
    "url" : "https://github.com/keras-team/keras"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/releases/tag/v3.12.3"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/keras-team/keras/releases/tag/v3.15.0"
  }, {
    "type" : "WEB",
    "url" : "https://huntr.com/bounties/ab14df49-13b5-4442-b754-3189430bfa28"
  } ],
  "database_specific" : {
    "cwe_ids" : [ "CWE-502" ],
    "severity" : "HIGH",
    "github_reviewed" : true,
    "github_reviewed_at" : "2026-08-07T20:33:49Z",
    "nvd_published_at" : "2026-07-19T20:16:28Z"
  }
}