GHSA-36fq-jgmw-4r9c
Keras is vulnerable to Deserialization of Untrusted Data
Packages
Details
As written in the advisory (Markdown source, shown as plain text).
### Arbitrary Code Execution in Keras
Keras versions prior to 3.11.0 allow for arbitrary code execution when loading a crafted `.keras` model archive, even when `safe_mode=True`.
The issue arises because the archive’s `config.json` is parsed before layer deserialization. This can invoke `keras.config.enable_unsafe_deserialization()`, effectively disabling safe mode from within the loading process itself. An attacker can place this call first in the archive and then include a `Lambda` layer whose function is deserialized from a pickle, leading to the execution of attacker-controlled Python code as soon as a victim loads the model file.
Exploitation requires a user to open an untrusted model; no additional privileges are needed. The fix in version 3.11.0 enforces safe-mode semantics *before* reading any user-controlled configuration and prevents the toggling of unsafe deserialization via the config file.
**Affected versions:** < 3.11.0
**Patched version:** 3.11.0
It is recommended to upgrade to version 3.11.0 or later and to avoid opening untrusted model files.
References
External links from the advisory. They lead to third-party sites.
- https://nvd.nist.gov/vuln/detail/CVE-2025-9906
- https://github.com/keras-team/keras/pull/21429
- https://github.com/keras-team/keras/commit/713172ab56b864e59e2aa79b1a51b0e728bba858
- https://github.com/keras-team/keras
- https://github.com/keras-team/keras/releases/tag/v3.11.0
- https://github.com/pypa/advisory-database/tree/main/vulns/keras/PYSEC-2025-76.yaml
- https://osv.dev/vulnerability/CVE-2025-9906