CVE-2026-43627: llama.cpp b1283–b9058 Integer Overflow in llama_batch_init() Function
llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llamabatchinit() function where unchecked multiplications in malloc() calls can wrap past INT32MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries.
Affected Software
Event History
Frequently Asked Questions
What is the severity of CVE-2026-43627?
The severity of CVE-2026-43627 is rated as high with a score of 7.8.
How do I fix CVE-2026-43627?
To fix CVE-2026-43627, update to the latest version of llama.cpp that includes the security patches addressing the integer overflow vulnerability.
What types of vulnerabilities does CVE-2026-43627 exploit?
CVE-2026-43627 exploits an integer overflow vulnerability in the llama_batch_init() function.
What could an attacker achieve by exploiting CVE-2026-43627?
An attacker could potentially perform unauthorized actions by triggering an integer overflow that affects memory allocation.
What software is affected by CVE-2026-43627?
The software affected by CVE-2026-43627 is llama.cpp, specifically builds from b1283 through b9058.