CVE-2026-70639: llama.cpp b1886–b7445 Null Pointer Dereference DoS via llama-android.cpp

Published Aug 6, 2026
·
Updated

llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service.

Affected Software

1 affected component
llama.cpp>=b1886<=b7445

Event History

Aug 6, 2026
CVE Published
via MITRE·03:37 PM
Data Sourced
via MITRE·03:37 PM
DescriptionSeverityWeakness
Data Sourced
via NVD·10:18 PM
DescriptionSeverityWeakness
Free Weekly Intel

Don't miss critical vulnerabilities

Join thousands of security professionals who receive our weekly digest of trending CVEs, zero-days, and exploited vulnerabilities.

No spam. Unsubscribe anytime.

Frequently Asked Questions

1

What is the severity of CVE-2026-70639?

CVE-2026-70639 has a medium severity rating of 5.5.

2

How do I fix CVE-2026-70639?

To fix CVE-2026-70639, update to version b7446 or later of the llama.cpp software.

3

What impact does CVE-2026-70639 have on the application?

CVE-2026-70639 can lead to a denial of service (DoS) attack through a null pointer dereference.

4

What software is affected by CVE-2026-70639?

The vulnerability affects versions b1886 through b7445 of the llama.cpp software.

5

How can attackers exploit CVE-2026-70639?

Attackers can exploit CVE-2026-70639 by supplying a malicious, corrupt, or truncated model file.

Contact

SecAlerts Pty Ltd.
132 Wickham Terrace
Fortitude Valley,
QLD 4006, Australia
info@secalerts.co
By using SecAlerts services, you agree to our services end-user license agreement. This website is safeguarded by reCAPTCHA and governed by the Google Privacy Policy and Terms of Service. All names, logos, and brands of products are owned by their respective owners, and any usage of these names, logos, and brands for identification purposes only does not imply endorsement. If you possess any content that requires removal, please get in touch with us.
© 2026 SecAlerts Pty Ltd.
ABN: 70 645 966 203, ACN: 645 966 203