Vulnerability Monitor

The vendors, products, and vulnerabilities you care about

CVE-2026-70639


llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() 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.


Security Impact Summary

This vulnerability carries a MEDIUM severity rating with a CVSS v3.1 score of 5.5, requiring local system access to exploit with relatively low complexity though user interaction is required and does not require pre-existing privileges . The vulnerability impacts and availability (service disruption) for affected systems.

Historical Context

Reported in 2026, this vulnerability emerged during an era marked by increased sophistication in supply chain attacks, cloud infrastructure vulnerabilities, and software-as-a-service (SaaS) security challenges. Security practices during this period emphasized zero-trust architectures, container security, and API protection.


Published

2026-08-06T22:18:28.783

Last Modified

2026-08-31T20:31:19.777

Status

Undergoing Analysis

Source

[email protected]

Severity

CVSSv3.1: 5.5 (MEDIUM)

Weaknesses
  • Type: Secondary
    CWE-476

Affected Vendors & Products

-


References

How SecUtils Interprets This CVE

SecUtils normalizes and enriches National Vulnerability Database (NVD) records by standardizing vendor and product identifiers, aggregating vulnerability metadata from both NVD and MITRE sources, and providing structured context for security teams. For affected software, we extract Common Platform Enumeration (CPE) data, Common Weakness Enumeration (CWE) classifications, CVSS severity metrics, and reference data to enable rapid vulnerability prioritization and asset correlation. This record contains no exploit code, proof-of-concept instructions, or attack methodologies—only defensive intelligence necessary for patch management, risk assessment, and security operations.