llama.cpp: Llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android…
llama.cpp
More for llama.cpp RSS feed for llama.cpp
- Severity
- High, CVSS 7.8
- CVSS vector
- CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
- Record
- CVE-2026-43622
- Affects
- llama.cpp
- Known exploited
- Not listed in the CISA Known Exploited Vulnerabilities catalog
- Published
- Record first published by the source
- August 6, 2026
- Source last modified
- October 23, 2026
Description
llama.cpp builds b1886 through b7445 contain a double free vulnerability in the LLaMA-Android JNI wrapper where new_1batch() allocates memory using malloc() while free_1batch() deallocates it using the C++ delete operator, causing heap metadata corruption. Attackers can trigger this memory management mismatch to cause denial of service through process crashes or potentially achieve arbitrary code execution depending on allocator state.
Description quoted from the NVD record.
Who is affected
This flaw affects llama.cpp, made by ggml, which is software for running AI language models on your own computers, servers or devices. It affects builds b1886 through b7445. The problem is in the part of the product that lets Android apps use llama.cpp. A mistake in how that part handles memory means an attacker can make the app crash so it stops working. In some cases the attacker may also be able to run their own code on the device. This flaw is rated High.
What to do
- If someone in your organization built or runs an Android app that uses llama.cpp, ask that person to check which llama.cpp build the app includes. Builds b1886 through b7445 are affected.
- No fixed version is named for this flaw yet, so check the vendor advisory linked on this page for a fixed release, and have the app updated to it once one is available.
- If an IT provider or vendor makes or manages the app for your organization, send them this advisory and ask when the update will be done.
Analysis written from the source record as of October 23, 2026 and reviewed before publishing.