You can deploy this model and start earning money today!

TheAuroraAi/
keras-h5-oom-poc

Name: keras-h5-oom-poc
Availability: InStock

Views

No views yet

CWE-789: Memory Allocation via H5 Shape Inflation in Keras

A crafted .keras file (~1.2 KB) triggers 4.75 GB allocation via load_model(). H5 dataset shapes bypass is_memory_sufficient() check.

bash

pip install keras tensorflow h5py
python3 keras_oom_poc.py

ValueError: Unable to allocate 4.00 GiB for array with shape (300000, 3579) OR: Process killed by OOM killer

1,239 bytes input -> 4,750,000,000 bytes allocated (3.8M x ratio)

saving_lib.py:1176 — np.array(value) trusts H5 dataset shape. is_memory_sufficient() checks MODEL shapes, not H5 shapes.

CWE-789: Memory Allocation with Excessive Size Value