CVE-2025-9906: keras-team keras Deserialization of Untrusted Data

Vulnerability summary

CVE-2025-9906 is a high (CVSS 8.6) Deserialization of Untrusted Data vulnerability in keras-team keras. EPSS estimates a 0.20% chance of exploitation in the next 30 days, higher than 9.6% of scored vulnerabilities.

Timeline

1. 19-09-2025
   PublishedVulnerability first disclosed.
2. 26-02-2026
   Last modifiedVulnerability information updated.

Description

The Keras Model.load_model method can be exploited to achieve arbitrary code execution, even with safe_mode=True. One can create a specially crafted .keras model archive that, when loaded via Model.load_model, will trigger arbitrary code to be executed. This is achieved by crafting a special config.json (a file within the .keras archive) that will invoke keras.config.enable_unsafe_deserialization() to disable safe mode. Once safe mode is disable, one can use the Lambda layer feature of keras, which allows arbitrary Python code in the form of pickled code. Both can appear in the same archive. Simply the keras.config.enable_unsafe_deserialization() needs to appear first in the archive and the Lambda with arbitrary code needs to be second.

• CWE-502Deserialization of Untrusted DataThe product deserializes untrusted data without sufficiently ensuring that the resulting data will be valid.