High severity · published Aug 6, 2026
CVE-2026-43631
llama.cpp b7492–b9060 Use-After-Free RCE via llama-server
8.1
CVSS
0.6%
EPSS
No
On KEV
Description
llama.cpp builds b7492 through the latest b9060 contains a use-after-free vulnerability in the vocab pointer of llama-server when the --sleep-idle-seconds feature is enabled, allowing unauthenticated remote attackers to execute arbitrary code. Attackers can trigger the vulnerability by sending requests to affected endpoints while the server transitions to sleep mode, causing concurrent worker threads to dereference a freed vocab pointer that can be reclaimed with attacker-controlled data to achieve remote code execution.
CVSS vector
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
Attack vector
Network
Complexity
High
Privileges
None
User interaction
None
Confidentiality
High
Integrity
High
Availability
High
Scope
Unchanged
Exploitation
Is it being used?
Exploit prediction (EPSS)0.6% · higher than 47% of CVEs
Exploited in the wildNo evidence yet
CISA KEVNot listed
Ransomware campaign useNot known
Affected products
Which versions
LL
llama.cpp · ggml-org
b7492 through b9060 · source: vendor, NVD
References
NVD
https://nvd.nist.gov/vuln/detail/CVE-2026-43631
CVE.org
https://www.cve.org/CVERecord?id=CVE-2026-43631
Patch
https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches
Patch
https://www.vulncheck.com/advisories/llama-cpp-b7492-b9060-use-after-free-rce-via-llama-server
More from llama.cpp
9.8
CVE-2026-34159
llama.cpp: Unauthenticated RCE via GRAPH_COMPUTE buffer=0 bypass in llama.cpp RPC backend
9.8
CVE-2026-21869
llama.cpp has Out-of-bounds Write in llama-server
8.9
CVE-2025-53630
Integer Overflow in GGUF Parser can lead to Heap Out-of-Bounds Read/Write in gguf
8.8
CVE-2025-52566
llama.cpp tokenizer signed vs. unsigned heap overflow
8.8
CVE-2025-49847
llama.cpp Vulnerable to Buffer Overflow via Malicious GGUF Model
