CVE-2024-42479
llama.cpp provides LLM inference in C/C++. The unsafe `data` pointer member in the `rpc_tensor` structure can cause arbitrary address writing. This vulnerability is fixed in b3561.
Scoring
- Severity
- CRITICAL
- CVSS base score
- 10
- CVSS vector
- CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
- EPSS probability
- 2.63%
- CWE
- CWE-123
- Published
- 2024-08-12
- Last modified
- 2026-03-13
Affected products
- ggerganov llama.cpp
Weakness type
Related vulnerabilities
- CVE-2026-81579 — An untrusted Pointer Dereference can be exploited to escalate privileges by an unprivileged user on Windows
- CVE-2026-20469 — In trusted_mem, there is a possible escalation of privilege due to improper input validation. This...
- CVE-2026-47473 — NVIDIA TensorRT-LLM contains a vulnerability where an attacker could cause a write-what-where...
- CVE-2026-45257 — Arbitrary file overwrite via the KTLS receive path
- CVE-2026-41952 — Local privilege escalation due to improper input validation. The following products are affected:...
- CVE-2025-14857 — Semtech LR11xx Memory Write Access Control Bypass
- CVE-2026-25634 — iccDEV memcpy-param-overlap in CIccTagMultiProcessElement::Apply()
- CVE-2025-29943 — Write what were condition within AMD CPUs may allow an admin-privileged attacker to modify the...