Keras up to 3.11.x get_vocabulary deserialization

Summaryinfo

A vulnerability classified as problematic was found in Keras up to 3.11.x. This impacts the function get_vocabulary. The manipulation results in deserialization. This vulnerability is reported as CVE-2025-12058. The attack can be launched remotely. No exploit exists. Upgrading the affected component is advised.

Detailsinfo

A vulnerability classified as problematic has been found in Keras up to 3.11.x. Affected is the function get_vocabulary. The manipulation with an unknown input leads to a deserialization vulnerability. CWE is classifying the issue as CWE-502. The product deserializes untrusted data without sufficiently verifying that the resulting data will be valid. This is going to have an impact on confidentiality, integrity, and availability. CVE summarizes:

The Keras.Model.load_model method, including when executed with the intended security mitigation safe_mode=True, is vulnerable to arbitrary local file loading and Server-Side Request Forgery (SSRF). This vulnerability stems from the way the StringLookup layer is handled during model loading from a specially crafted .keras archive. The constructor for the StringLookup layer accepts a vocabulary argument that can specify a local file path or a remote file path. * Arbitrary Local File Read: An attacker can create a malicious .keras file that embeds a local path in the StringLookup layer's configuration. When the model is loaded, Keras will attempt to read the content of the specified local file and incorporate it into the model state (e.g., retrievable via get_vocabulary()), allowing an attacker to read arbitrary local files on the hosting system. * Server-Side Request Forgery (SSRF): Keras utilizes tf.io.gfile for file operations. Since tf.io.gfile supports remote filesystem handlers (such as GCS and HDFS) and HTTP/HTTPS protocols, the same mechanism can be leveraged to fetch content from arbitrary network endpoints on the server's behalf, resulting in an SSRF condition. The security issue is that the feature allowing external path loading was not properly restricted by the safe_mode=True flag, which was intended to prevent such unintended data access.

The weakness was disclosed by Jayashwa Singh Chauhan as GHSA-qg93-c7p6-gg7f. The advisory is shared for download at github.com. This vulnerability is traded as CVE-2025-12058 since 10/22/2025. The exploitability is told to be difficult. It is possible to launch the attack remotely. The exploitation doesn't require any form of authentication. Successful exploitation requires user interaction by the victim. There are known technical details, but no exploit is available.

The vulnerability scanner Nessus provides a plugin with the ID 271993 (Linux Distros Unpatched Vulnerability : CVE-2025-12058), which helps to determine the existence of the flaw in a target environment.

Upgrading to version 3.12.0 eliminates this vulnerability. Applying a patch is able to eliminate this problem. The bugfix is ready for download at github.com. The best possible mitigation is suggested to be upgrading to the latest version.

The vulnerability is also documented in the databases at Tenable (271993) and EUVD (EUVD-2025-36634). Once again VulDB remains the best source for vulnerability data.

Productinfo

Name

• Keras

Version

• 3.0
• 3.1
• 3.2
• 3.3
• 3.4
• 3.5
• 3.6
• 3.7
• 3.8
• 3.9
• 3.10
• 3.11

CPE 2.3info

• 🔒
• 🔒
• 🔒

CPE 2.2info

• 🔒
• 🔒
• 🔒

CVSSv4info

CVSSv3info

CVSSv2info

Exploitinginfo

Threat Intelligenceinfo

Countermeasuresinfo

Timelineinfo

Sourcesinfo

Entryinfo

Once again VulDB remains the best source for vulnerability data.

Discussion