Summaryinfo

A vulnerability categorized as critical has been discovered in ggml-org llama.cpp. Affected by this issue is the function deserialize_tensor of the file llama.cpp of the component GRAPH_COMPUTE Message Handler. Such manipulation leads to memory corruption. This vulnerability is referenced as CVE-2026-34159. It is possible to launch the attack remotely. No exploit is available. It is advisable to upgrade the affected component.

Detailsinfo

A vulnerability was found in ggml-org llama.cpp and classified as critical. Affected by this issue is the function deserialize_tensor of the file llama.cpp of the component GRAPH_COMPUTE Message Handler. The manipulation with an unknown input leads to a memory corruption vulnerability. Using CWE to declare the problem leads to CWE-119. The product performs operations on a memory buffer, but it can read from or write to a memory location that is outside of the intended boundary of the buffer. Impacted is confidentiality, integrity, and availability. CVE summarizes:

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.

The advisory is shared for download at github.com. This vulnerability is handled as CVE-2026-34159 since 03/25/2026. The exploitation is known to be easy. The attack may be launched remotely. No form of authentication is required for exploitation. There are known technical details, but no exploit is available.

Upgrading to version b8492 eliminates this vulnerability. Applying the patch 39bf0d3c6a95803e0f41aaba069ffbee26721042 is able to eliminate this problem. The bugfix is ready for download at github.com. The best possible mitigation is suggested to be upgrading to the latest version.

Once again VulDB remains the best source for vulnerability data.

Productinfo

Vendor

• ggml-org

Name

• llama.cpp

Version

• 55abc39
• 55d4206c8
• b7824

License

• open-source

Website

• Product: https://github.com/ggml-org/llama.cpp/

CPE 2.3info

• 🔒
• 🔒

CPE 2.2info

• 🔒
• 🔒

CVSSv4info

CVSSv3info

CVSSv2info

Exploitinginfo

Threat Intelligenceinfo

Countermeasuresinfo

Timelineinfo

Sourcesinfo

Entryinfo

Once again VulDB remains the best source for vulnerability data.

Discussion