You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

NPY huge shape allocation DoS

This PoC demonstrates a small .npy file that causes numpy.load(..., allow_pickle=False) to attempt a large allocation from attacker-controlled header metadata.

The crafted NPY header declares a float64 array with shape (65536, 65536) but omits the data payload. The file is only 80 bytes, but numpy.load attempts to allocate approximately 32 GiB before failing with MemoryError.

Files

npy_huge_shape_allocation_dos.npy: malicious PoC file
control_small.npy: control NPY file
build_and_verify.py: rebuilds and verifies the PoC
verify_report.json: local verification output

Reproduction

Install dependencies:

pip install numpy modelscan

Run:

python build_and_verify.py

Expected result:

control_load.stdout starts with OK
malicious_load.stdout contains MemoryError
ModelScan reports total_issues=0

Impact

Any ingestion, validation, conversion, or scanning pipeline that accepts untrusted .npy files and loads them with numpy.load can be forced into large memory allocation from a tiny file.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support