Home

Description

llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs. unsigned integer overflow in llama.cpp's tokenizer implementation (llama_vocab::tokenize) (src/llama-vocab.cpp:3036) resulting in unintended behavior in tokens copying size comparison. Allowing heap-overflowing llama.cpp inferencing engine with carefully manipulated text input during tokenization process. This issue has been patched in version b5721.

PUBLISHED Reserved 2025-06-18 | Published 2025-06-24 | Updated 2025-06-24 | Assigner GitHub_M




HIGH: 8.6CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

Problem types

CWE-119: Improper Restriction of Operations within the Bounds of a Memory Buffer

CWE-195: Signed to Unsigned Conversion Error

Product status

< b5721
affected

References

github.com/...ma.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx exploit

github.com/...ma.cpp/security/advisories/GHSA-7rxv-5jhh-j6xx

github.com/...ommit/dd6e6d0b6a4bbe3ebfc931d1eb14db2f2b1d70af

cve.org (CVE-2025-52566)

nvd.nist.gov (CVE-2025-52566)

Download JSON