PatchTheCVECVE-2026-34159: llama.cpp is an inference of several LLM models in C/C++
CVE-2026-34159 is a critical-severity flaw (CVSS 9.8) affecting llama.cpp. Below: why it ranks where it does, which versions are affected, what to do, and the exact commands.
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 acc
Why this priority
- Critical severity
- Public exploit reference
- CISA SSVC: automatable with total technical impact
- Remotely exploitable, no login, no user action
A fixed release or distro patch exists.
Risk facts
| CVSS score | 9.8 (Critical) |
| Attack vector | Network (remote) |
| Attack complexity | Low |
| Privileges needed | None |
| User interaction | None |
| EPSS exploit chance | 1.6% (top 24%) |
| CISA exploited list | No |
| Weakness | CWE-119 |
| Published | 2026-04-01 |
| Last modified | 2026-10-26 |
Last updated 2026-10-26 · Sources: CVE Program, NVD, CISA KEV, EPSS and vendor advisories · How we score
Am I affected?
Timeline
- Published CVE record published
- Updated Record last modified
Exact fixes by distribution
Public exploits and detection
Public code exists for this CVE. Treat unpatched systems as at risk, and only run PoCs in a lab.
- GitHub PoC: casp3r0x0/CVE-2026-34159 (28 stars)
- GitHub PoC: rohithronanki/CVE-2026-34159-Vulnerability-Research-Analysis-Detection (0 stars)
- GitHub PoC: TeamN4C/SG-2026-0021 (0 stars)
- Write-up: github.com/ggml-org/llama.cpp/security/advisories/GHSA-j8rj-fmpv-wcxw
Scores and fixes here are computed automatically from those and other public sources. Confirm with the vendor advisory before changing production systems.
Quick answers
How do I fix CVE-2026-34159?
llama.cpp: upgrade to b8492. Exact commands for each OS are on this page.
Is CVE-2026-34159 being exploited?
CVE-2026-34159 is not on the CISA Known Exploited Vulnerabilities list at the time of the last update. Check again, as this can change.
What does CVE-2026-34159 affect?
CVE-2026-34159 affects llama.cpp. Severity is critical (CVSS 9.8). 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
How urgent is CVE-2026-34159?
Priority 66/100. Critical severity; Public exploit reference; CISA SSVC: automatable with total technical impact; Remotely exploitable, no login, no user action
Software pages
OpenAI and LLM tools vulnerabilities
Related CVEs
- CVE-2026-70640 HIGH 7.0 UPGRADE llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Androi
- CVE-2026-70638 HIGH 7.8 UPGRADE llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrappe
- CVE-2026-70639 MEDIUM 5.5 UPGRADE llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI
- CVE-2026-43629 HIGH 8.1 UPGRADE llama.cpp builds b4882 through b9058 contain a heap buffer overflow vulnerability in the KV cache state restor
- CVE-2026-43631 HIGH 8.1 UPGRADE llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer o