CVE-2026-34159

Published: Apr 01, 2026 Last Modified: Apr 01, 2026
ExploitDB:
Other exploit source:
Google Dorks:
CRITICAL 9,8
Source: [email protected]
Attack Vector: network
Attack Complexity: low
Privileges Required: none
User Interaction: none
Scope: unchanged
Confidentiality: high
Integrity: high
Availability: high

Description

AI Translation Available

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.

119

Improper Restriction of Operations within the Bounds of a Memory Buffer

Stable
Abstraction: Class Structure: Simple
Common Consequences
Security Scopes Affected:
Integrity Confidentiality Availability
Potential Impacts:
Execute Unauthorized Code Or Commands Modify Memory Read Memory Dos: Crash, Exit, Or Restart Dos: Resource Consumption (Cpu) Dos: Resource Consumption (Memory)
Applicable Platforms
Languages: Assembly, C, C++, Memory-Unsafe
Likelihood of Exploit: High
View CWE Details
https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a9580…
https://github.com/ggml-org/llama.cpp/commit/39bf0d3c6a95803e0f41aaba069ffbee26…
https://github.com/ggml-org/llama.cpp/pull/20908
https://github.com/ggml-org/llama.cpp/pull/20908
https://github.com/ggml-org/llama.cpp/security/advisories/G…
https://github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw