CVE-2026-33298 PUBLISHED

llama.cpp has a Heap Buffer Overflow via Integer Overflow in GGUF Tensor Parsing

Assigner: GitHub_M
Reserved: 18.03.2026 Published: 24.03.2026 Updated: 24.03.2026

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.

Metrics

CVSS Vector: CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
CVSS Score: 7.8

Product Status

Vendor ggml-org
Product llama.cpp
Versions
  • Version < b7824 is affected

References

Problem Types

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