Hugging Face's logo

codeconflict7
/
gguf-ggmlpad-poc

GGUF
security-research
Model card Files
xet
 Community

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.

Log in or Sign Up to review the conditions and access this model content.

GGML_PAD Integer Overflow PoC (GGUF)

Security research PoC — DO NOT USE AS A REAL MODEL.

Files

  • poc_ggmlpad_overflow.gguf — 256-byte crafted GGUF v3 file
  • poc_ggmlpad_overflow.py — Python script to reproduce

Reproduce 

pip install llama-cpp-python numpy
python poc_ggmlpad_overflow.py

What it does

The GGUF file contains a tensor with ne[0] = 2^63 - 3, causing GGML_PAD to overflow to 0. This bypasses the CVE-2025-53630 overflow check, resulting in an undersized allocation and out-of-bounds tensor data pointer.

Downloads last month
-
GGUF
Model size
9223372T params
Architecture
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

View all variants
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support