Instructions to use srvrX0r/poc-keras-unsafe-deserialization with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use srvrX0r/poc-keras-unsafe-deserialization with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://srvrX0r/poc-keras-unsafe-deserialization") - Notebooks
- Google Colab
- Kaggle
PoC: Code Execution via Unsafe Keras Deserialization
This .keras model demonstrates arbitrary code execution during loading when safe_mode=False is used — a common misconfiguration in ML pipelines.
⚠️ For research and validation only. Do not load in production.
Payload Behavior
On load, the model executes:
echo 'SCANNER-BUYPass' > /tmp/pwned
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support