CVE-2026-70638: llama.cpp b1886–b7445 Integer Overflow via new_1batch() in llama-android.cpp
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new1batch() function multiplies sizeof(llamaseqid) by an attacker-controlled nseqmax parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted nseqmax value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
Affected Software
Event History
Frequently Asked Questions
What is the severity of CVE-2026-70638?
The severity of CVE-2026-70638 is rated high with a score of 7.8.
What is the risk associated with CVE-2026-70638?
CVE-2026-70638 has a risk rating of 68, indicating significant potential for exploitation.
How do I fix CVE-2026-70638?
To fix CVE-2026-70638, upgrade to versions of llama.cpp beyond b7445 which contain the necessary patches.
What type of vulnerability is CVE-2026-70638?
CVE-2026-70638 is categorized as an integer overflow vulnerability.
Which function is affected in CVE-2026-70638?
The integer overflow vulnerability in CVE-2026-70638 is found in the new_1batch() function in llama-android.cpp.