Views
No views yet
overflow_poc.gguf — 2-tensor variant (intra-allocation OOB)overflow_poc_v2.gguf — Single-tensor variant (NULL deref, ASAN-confirmed)poc.py — Generator script with arithmetic proof1git clone https://github.com/ggerganov/llama.cpp && cd llama.cpp
2cmake -B build -DBUILD_SHARED_LIBS=OFF
3cmake --build build --target llama-gguf
4./build/bin/llama-gguf overflow_poc.gguf r n