Unauthenticated RCE via unchecked null-buffer tensor
Published Apr 1, 2026 · Updated Apr 1, 2026
Memory corruption in the RPC backend in ggml.ai llama.cpp before b8492 allows remote attackers to execute arbitrary code via TCP. The deserialize_tensor() path accepts an attacker-supplied data pointer when buffer is zero, bypassing bounds validation before compute kernels use it. Exploitation requires TCP access to an explicitly enabled RPC server and combines protocol pointer leaks with arbitrary memory reads and writes to execute code as the server process.
Summary
What happened
Memory corruption in the RPC backend in ggml.ai llama.cpp before b8492 allows remote attackers to execute arbitrary code via TCP. The deserialize_tensor() path accepts an attacker-supplied data pointer when buffer is zero, bypassing bounds validation before compute kernels use it. Exploitation requires TCP access to an explicitly enabled RPC server and combines protocol pointer leaks with arbitrary memory reads and writes to execute code as the server process.
The record
- CVE
- CVE-2026-34159
- Published
- Apr 1, 2026
- Updated
- Apr 1, 2026
- Vendor
- Georgi Gerganov
- Product
- llama.cpp
- Classifications
- CWE-119, T1210
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Apr 1, 2026CVE publishedPublication date reported by the CVE source.
- Apr 1, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< b8492
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- GHSA-j8rj-fmpv-wcxw RPC RCE proof of conceptfunctional · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo