Vulnerability · Published Sep 19, 2025 · Assigned by Google
CVE-2025-9905Arbitary Code execution in Keras load_model()
High 7.3
In plain English
A high-severity security flaw in Keras from Keras-team. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 13% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.
Official 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 .h5/.hdf5 model archive that, when loaded via Model.load_model, will trigger arbitrary code to be executed. This is achieved by crafting a special .h5 archive file that uses the Lambda layer feature of keras which allows arbitrary Python code in the form of pickled code. The vulnerability comes from the fact that the safe_mode=True option is not honored when reading .h5 archives. Note that the .h5/.hdf5 format is a legacy format supported by Keras 3 for backwards compatibility.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| PyPI | keras | 3.11.3 | GHSA-36rr-ww3j-vrjv |
| PyPI | keras | 3.11.3 | PYSEC-2025-123 |
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| Keras-team | Keras | 3.0.0 – through 3.11.2 |
Weakness: CWE-913 · Vector: CVSS:4.0/AV:L/AC:H/AT:P/PR:L/UI:P/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H