Load-Time Division-by-Zero Crash in llama.cpp GGUF Metadata Parser

โš ๏ธ This repository does not contain a machine learning model.
The file provided is intentionally malformed and exists solely for security research and vulnerability reproduction.

PoC Artifact

  • poc_fp_exception.gguf

Affected Component

  • llama.cpp โ€” GGUF metadata parser

Vulnerability Summary

A malformed GGUF file can trigger a deterministic crash during model loading due to a division-by-zero condition in GGUF tensor shape validation.

The crash occurs at load time, before inference or tensor data processing, and results in immediate process termination.

Reproduction

From a llama.cpp build directory:

./llama-gguf poc_fp_exception.gguf r
Downloads last month
27
GGUF
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support