CVE-2026-70640

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.

Weakness classification

• disclosure@vulncheck.com: CWE-362

Primary references

• disclosure@vulncheck.com — Patch, Third Party Advisory
• disclosure@vulncheck.com — Patch
• disclosure@vulncheck.com — Patch, Release Notes, Vendor Advisory
• disclosure@vulncheck.com — Exploit, Third Party Advisory