View at Official debian advisory, NVD, CVE.org · CVE detail
Freshness: upstream tracker timestamp is available; use API updated time as primary recency signal.
CVE-2025-49847 not yet assigned priority: Debian including 1 source packages (llama.cpp), 2 status rows across 2 suites (forky, sid): resolved 2.
llama.cpp is an inference of several LLM models in C/C++. Prior to version b5662, an attacker‐supplied GGUF model vocabulary can trigger a buffer overflow in llama.cpp’s vocabulary‐loading code. Specifically, the helper _try_copy in llama.cpp/src/vocab.cpp: llama_vocab::impl::token_to_piece() casts a very large size_t token length into an int32_t, causing the length check (if (length < (int32_t)size)) to be bypassed. As a result, memcpy is still called with that oversized size, letting a malicious model overwrite memory beyond the intended buffer. This can lead to arbitrary memory corruption and potential code execution. This issue has been patched in version b5662.