CVE-2026-34159

Name	CVE-2026-34159
Description	llama.cpp is an inference of several LLM models in C/C++. Prior to version b8492, the RPC backend's deserialize_tensor() skips all bounds validation when a tensor's buffer field is 0. An unauthenticated attacker can read and write arbitrary process memory via crafted GRAPH_COMPUTE messages. Combined with pointer leaks from ALLOC_BUFFER/BUFFER_GET_BASE, this gives full ASLR bypass and remote code execution. No authentication required, just TCP access to the RPC server port. This issue has been patched in version b8492.
Source	CVE (at NVD; CERT, ENISA, LWN, oss-sec, fulldisc, Debian ELTS, Red Hat, Ubuntu, Gentoo, SUSE bugzilla/CVE, GitHub advisories/code/issues, web search, more)

Vulnerable and fixed packages

The table below lists information on source packages.

Source Package	Release	Version	Status
llama.cpp (PTS)	forky, sid	8681+dfsg-1	fixed

The information below is based on the following data on fixed versions.

Package	Type	Release	Fixed Version	Urgency	Origin	Debian Bugs
llama.cpp	source	(unstable)	8611+dfsg-1

Notes

https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw
https://github.com/ggml-org/llama.cpp/pull/20908
Fixed by: https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042 (b8492)