Vulnerability · Published Apr 1, 2026 · Assigned by GitHub_M
CVE-2026-34159llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
Critical 9.8
In plain English
A critical-severity memory corruption in llama.cpp from ggml-org. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 1.6% (higher than 76% of all known vulnerabilities). Ready-made attack tooling exists: public proof-of-concept code. Fixed versions are available - upgrade the affected package.
Official 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.
Affected packages & fixed versions
| Ecosystem | Package | Fixed in | Advisory |
|---|---|---|---|
| Debian:14 | llama.cpp | 8611+dfsg-1 | DEBIAN-CVE-2026-34159 |
| Ubuntu:25.10 | llama.cpp | no fix listed | UBUNTU-CVE-2026-34159 |
Affected products (as reported)
Weakness: CWE-119 · Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
References
- https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw x_refsource_CONFIRM
- https://github.com/ggml-org/llama.cpp/pull/20908 x_refsource_MISC
- https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042 x_refsource_MISC