CVEvulnerability

CVE-2026-33298

7.8 HIGH Published 2026-03-24 Modified 2026-06-17 Official page ↗
Status
Analyzed
Published
2026-03-24
Last modified
2026-06-17
Assigner
[email protected]
CVSS 3.1 ([email protected])
7.8 HIGH — CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Weaknesses
CWE-122, CWE-190

Description

llama.cpp is an inference of several LLM models in C/C++. Prior to b7824, an integer overflow vulnerability in the `ggml_nbytes` function allows an attacker to bypass memory validation by crafting a GGUF file with specific tensor dimensions. This causes `ggml_nbytes` to return a significantly smaller size than required (e.g., 4MB instead of Exabytes), leading to a heap-based buffer overflow when the application subsequently processes the tensor. This vulnerability allows potential Remote Code Execution (RCE) via memory corruption. b7824 contains a fix.

Affected configurations

• ggml llama.cpp (< b7824)

Related

Weakness · CWE (2)

  • CWE-122 Heap-based Buffer Overflow weakness
  • CWE-190 Integer Overflow or Wraparound weakness

References