CVE-2026-70638
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
वीकनेस
llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding.
प्रभावित प्रोडक्ट
- ggml llama.cpp
संदर्भ
- https://github.com/Vladimir-tokarev-cyera/llama-cpp-security-patches
- https://github.com/ggml-org/llama.cpp/commit/5c0d18881e0e9794c96b2602736b758bac9d9388
- https://github.com/ggml-org/llama.cpp/releases/tag/b7446
- https://www.vulncheck.com/advisories/llama-cpp-b1886-b7445-integer-overflow-via-new-1batch-in-llama-android-cpp
हमलावर से पहले बग आप ढूँढें।
GitHub से साइन इन करें और एक मिनट से भी कम में अपना पहला ऑडिट चलाएँ। फ़्री प्लान के लिए क्रेडिट कार्ड की ज़रूरत नहीं।