CVE-2024-58399
Published:October 27, 2026
Updated:October 28, 2026
llama.cpp before b2731 contains an out-of-bounds read vulnerability in llama_model_loader that allows attackers to read memory outside the mapped model file via crafted GGUF tensor offsets. Attackers can supply a crafted or truncated GGUF model file with arbitrary 64-bit tensor offsets to crash the process or load adjacent process memory as model weights.
Affected Packages
llama-cpp (CONAN):
Affected version(s) =b2038 <b3040Fix Suggestion:
Update to version b3040llama.cpp (CONDA):
Affected version(s) >=0.0.1601 <5289Fix Suggestion:
Update to version 5289https://github.com/ggml-org/llama.cpp.git (GITHUB):
Affected version(s) >=b1046 <b2731Fix Suggestion:
Update to version b2731Related Resources (8)
Do you need more information?
Contact UsCVSS v4
Base Score:
6.9
Attack Vector
LOCAL
Attack Complexity
LOW
Attack Requirements
NONE
Privileges Required
NONE
User Interaction
PASSIVE
Vulnerable System Confidentiality
LOW
Vulnerable System Integrity
NONE
Vulnerable System Availability
HIGH
Subsequent System Confidentiality
NONE
Subsequent System Integrity
NONE
Subsequent System Availability
NONE
CVSS v3
Base Score:
6.1
Attack Vector
LOCAL
Attack Complexity
LOW
Privileges Required
NONE
User Interaction
REQUIRED
Scope
UNCHANGED
Confidentiality
LOW
Integrity
NONE
Availability
HIGH
Weakness Type (CWE)
Out-of-bounds Read
EPSS
Base Score:
0.12