CVE-2025-53630

LLM models - Memory Corruption

Title source: llm

Description

llama.cpp is an inference of several LLM models in C/C++. Integer Overflow in the gguf_init_from_file_impl function in ggml/src/gguf.cpp can lead to Heap Out-of-Bounds Read/Write. This vulnerability is fixed in commit 26a48ad699d50b6268900062661bd22f3e792579.

References (2)

[Patch] https://github.com/ggml-org/llama.cpp/commit/26a48ad699d50b6268900062661bd22f3e792579

View Patch ZIP pw:eip

[Vendor Advisory] https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-vgg9-87g3-85w8

Scores

EPSS 0.0006

EPSS Percentile 19.0%

Classification

CWE

CWE-122 CWE-680

Status draft

Timeline

Published Jul 10, 2025

Tracked Since Feb 18, 2026