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 (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

NumPy (.npy) Scanner Bypass PoC

Vulnerability Summary

Format: NumPy (.npy) Scanner Bypass: modelscan 0.8.7 (NumpyUnsafeOpScan) + picklescan 1.0.1 (DUAL BYPASS) Impact: Arbitrary Code Execution

Key Finding

modelscan ACTIVELY SCANS this file with NumpyUnsafeOpScan but still reports "No issues found." The NumpyUnsafeOpScan scanner uses pickletools.genops() to parse the pickle stream inside the file. The joblib numpy byte interleaving technique crashes the parser at position 235 before reaching the malicious reduce call.

Scanner Output

Scanning malicious_model.npy using modelscan.scanners.NumpyUnsafeOpScan model scan
--- Summary ---
No issues found!
--- Errors ---
Error 1: Parsing error: at position 235, opcode b'' unknown

Reproduction

modelscan -p malicious_model.npy    # NumpyUnsafeOpScan runs, crashes, reports clean
picklescan -p malicious_model.npy   # Crashes, reports clean
python3 -c "import joblib; joblib.load('malicious_model.npy')"  # ACE confirmed

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