CVE-2026-70640 | BDVDB-2026-356152 Detail
ggml-orgllama.cpp
Description#
llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution.
Impact#
Per CVSS v3.1: confidentiality impact high, integrity impact high, availability impact high. Attack vector: local.
Weaknesses#
References#
- https://github.com/ggml-org/llama.cpp/releases/tag/b7446
- https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388
- https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches
- https://www.vulncheck.com/advisories/llama-cpp-b1886-b7445-race-condition-use-after-free-via-llama-android-cpp
Exploit#
No public exploit known
History#
- Version 1Record published in the CVE List.
- Version 2Most recent update recorded in the CVE List.