Security Advisory

CVE-2026-27940

CVE vulnerability detail - eXtreme Datacenter Security Operations

Published 2026-03-12 16:39:37

Last updated 2026-03-14 03:55:24

Assigner GitHub_M

CVSS score 7.8

State PUBLISHED

Description

llama.cpp is an inference of several LLM models in C/C++. Prior to b8146, the gguf_init_from_file_impl() in gguf.cpp is vulnerable to an Integer overflow, leading to an undersized heap allocation. Using the subsequent fread() writes 528+ bytes of attacker-controlled data past the buffer boundary. This is a bypass of a similar bug in the same file - CVE-2025-53630, but the fix overlooked some areas. This vulnerability is fixed in b8146.

← Browse all CVEs View on cve.org ↗