What’s Vulnerable › Vulnerabilities

CVE-2026-34159: llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend

• Severity: critical (CVSS 9.8 v3.1)
• Exploited in the wild: not listed in CISA KEV
• EPSS: 1.6% probability of exploitation in the next 30 days
• Public exploit or detection template: yes
• Published: 2026-04-01 · updated 2026-10-26
• Weakness: CWE-119

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 products

• ggml-org llama.cpp: before b8492 — fixed in b8492

How to fix

Upgrade to a fixed version: ggml-org llama.cpp b8492. Check the vendor advisory for mitigations.

References

• https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26721042
• https://github.com/ggml-org/llama.cpp/pull/20908
• https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw

Get alerted when CVE-2026-34159 or the next vulnerability affects your stack